Author Topic: Safe Intantaneous Impact Point Tracks for Starship Recovery  (Read 20011 times)

Offline whitelancer64

The FAA did not regulate or license Shuttle flights. They would have done so if a private company had taken over Shuttle operations.

The Shuttle not only kept flying, but also kept overflying the CONUS during some reentry / landings. Notably, STS-116 (December 2006, the 4th flight after Columbia) overflew the Gulf coast, pretty much right over Houston and New Orleans, on the way to Florida.

I mentioned earlier that Dragon Crew-4 reentry was more or less directly overhead St. Louis, Nashville, and Atlanta. Apparently the risk is low enough for FAA / NASA to allow for overflight of highly populated areas during reentry.

To be clear, I am not saying you don't have a valid concern here, but I think the risk is probably much lower than you think (IIRC NASA required commercial crew operators to have a 1/500 LOC risk during reentry), and / or not calculated in the methods / terms you are using.
"One bit of advice: it is important to view knowledge as sort of a semantic tree -- make sure you understand the fundamental principles, ie the trunk and big branches, before you get into the leaves/details or there is nothing for them to hang on to." - Elon Musk
"There are lies, damned lies, and launch schedules." - Larry J

Offline whitelancer64

Aha. So the lede isn't buried:

https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_450.123-1_Population_Exposure_Assessment_2022.pdf

I am pretty sure this is the document you need to read.

An Advisory Circular has not yet been issued for what is also relevant, 14 CFR § 450.119 - Trajectory analysis for malfunction flight.

https://www.law.cornell.edu/cfr/text/14/450.119

I started looking through FAA training materials for the Part 450 regulations.

https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_1_Part_450_Workshop_Slides_Final.pdf
https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_2_Part_450_Workshop_Slides_Final.pdf
https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_3_Part_450_Workshop_Slides_Final.pdf

Day 2 talks about Flight Safety Analysis with respect to debris, ballistics, etc. and Acceptable Risk Limits. It DOES talk about instantaneous impact points, but exclusively in the context of launch and launch abort and associated downrange debris impacts. Population exposure risks are talked about on pages 60-62. This is very short in context of the document.

Almost everything discussed in the training materials is about launch and launch abort risks. It basically requires doing a massive amount of analysis to identify and assess risks and develop mitigations for them. Reentry is, as far as I can tell, exclusively mentioned as "launch or reentry" or "launch / reentry," presumably everything applies to reentry as it does to launch, but as far as I can tell, that's not specified. Reentry is just not talked about on its own.

There is a lot of safety criteria mentioned in Day 1 for different things, but generally:

Maximum individual risk is 1 in a million - land areas must be clear of people where the probability of a casualty is greater than 1 in a million, and Maximum collective risk is 1 in 10,000, such that 10,000 reentries would produce no more than 1 casualty on the ground.

"One bit of advice: it is important to view knowledge as sort of a semantic tree -- make sure you understand the fundamental principles, ie the trunk and big branches, before you get into the leaves/details or there is nothing for them to hang on to." - Elon Musk
"There are lies, damned lies, and launch schedules." - Larry J

Online sdsds

  • Senior Member
  • *****
  • Posts: 7106
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 1952
  • Likes Given: 1918
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #62 on: 12/06/2023 08:38 pm »
A few terminology/housekeeping points, intended to be helpful. All just my personal opinions or sense of things.

- Historically the acronym "IIP" has been used almost exclusively in the context of launch and ascent analyses, not entry, descent, and landing analyses.
- For calculations of public safety during EDL, the instantaneous debris field location (if a debris-generating event were to occur at that instant) seems like the important consideration.
- Regarding the curve an ascent IIP makes over time, the term used historically (and thus most useful for e.g. google searches) seems to be "IIP trace."
- Regarding the title of this thread, I would suggest something like "Public Safety During Starship Atmospheric Re-Entry."

« Last Edit: 12/07/2023 04:09 pm by sdsds »
— 𝐬𝐝𝐒𝐝𝐬 —

Offline Twark_Main

  • Senior Member
  • *****
  • Posts: 3577
  • Technically we ALL live in space
  • Liked: 1857
  • Likes Given: 1181
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #63 on: 12/06/2023 08:57 pm »
A few terminology/housekeeping points, intended to be helpful. All just my personal opinions or sense of things.

- Historically the acronym "IIP" has been used almost exclusively in the context launch and ascent analyses, not entry, descent, and landing analyses.

New history is being written daily.  :)

- For calculations of public safety during EDL, the instantaneous debris field location (if a debris-generating event were to occur at that instant) seems like the important consideration.

But not just the location of the debris mid-air, but mainly the Instantaneous location the debris would Impact the ground, which (as for launch) to model easily you might simplify to at least one Point.


- Regarding the title of this thread, I would suggest something like "Public Safety During Starship Atmospheric Re-Entry."

I would suggest it's fine.

These are helpful notes for Internet and literature searches, however. Thank you!
« Last Edit: 12/06/2023 09:04 pm by Twark_Main »
"The search for a universal design which suits all sites, people, and situations is obviously impossible. What is possible is well designed examples of the application of universal principles." ~~ David Holmgren

Online TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4324
  • Tampa, FL
  • Liked: 3243
  • Likes Given: 634
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #64 on: 12/07/2023 03:05 pm »
Aha. So the lede isn't buried:

https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_450.123-1_Population_Exposure_Assessment_2022.pdf

I am pretty sure this is the document you need to read.

An Advisory Circular has not yet been issued for what is also relevant, 14 CFR § 450.119 - Trajectory analysis for malfunction flight.

https://www.law.cornell.edu/cfr/text/14/450.119

I started looking through FAA training materials for the Part 450 regulations.

https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_1_Part_450_Workshop_Slides_Final.pdf
https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_2_Part_450_Workshop_Slides_Final.pdf
https://www.faa.gov/sites/faa.gov/files/space/streamlined_licensing_process/Day_3_Part_450_Workshop_Slides_Final.pdf

Day 2 talks about Flight Safety Analysis with respect to debris, ballistics, etc. and Acceptable Risk Limits. It DOES talk about instantaneous impact points, but exclusively in the context of launch and launch abort and associated downrange debris impacts. Population exposure risks are talked about on pages 60-62. This is very short in context of the document.

Almost everything discussed in the training materials is about launch and launch abort risks. It basically requires doing a massive amount of analysis to identify and assess risks and develop mitigations for them. Reentry is, as far as I can tell, exclusively mentioned as "launch or reentry" or "launch / reentry," presumably everything applies to reentry as it does to launch, but as far as I can tell, that's not specified. Reentry is just not talked about on its own.

There is a lot of safety criteria mentioned in Day 1 for different things, but generally:

Maximum individual risk is 1 in a million - land areas must be clear of people where the probability of a casualty is greater than 1 in a million, and Maximum collective risk is 1 in 10,000, such that 10,000 reentries would produce no more than 1 casualty on the ground.

The AC seems to be restricted just to computing population densities and structure/vehicle distributions on the ground, and doesn't handle the computation of the actual IIP track.  Some of that is in this AC_450.115-1A.

Note that 14 CFR Part 450 contains the common regulations for expendable vehicles (Part 415 and 417), reentry vehicles (Part 435) and reusable vehicles (Part 431).  I went through the reentry and reusable regs quickly and it doesn't appear that they call out reentry IIP tracks in any great detail.  I suspect they simply assume that the same kinds of methods described in AC-450.115-1A.

Note that, with the exception of Starship and the Shuttle, reentry vehicles are small.  If they break up, the debris tends to be small and of limited quantity.  That in itself will result in lower expected values.

Per your statement above, the Shuttle didn't have to go through this rigamarole.  IMO, if it had to pass the 14 CFR 450 regulations (which were finalized in 2020), it wouldn't have been able to.

Starship is roughly Shuttle-sized, but it has more engines, larger tanks, non-trivial, highly volatile prop residuals for landing, and it's made of a material with much tougher failure properties than aircraft aluminum.  I think you just have to crank through the debris-generating failure modes, use the specified ballistic propagators, convolve it with the population and structure data, and count your Dead People.  I don't see anything in either the regs or the AC's that makes entry fundamentally different from launch, other than the size of the launchers relative to that of the reentry vehicles.  In SH/SS's case, those relative size differences are smaller.

Online sdsds

  • Senior Member
  • *****
  • Posts: 7106
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 1952
  • Likes Given: 1918
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #65 on: 12/07/2023 04:36 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?
— 𝐬𝐝𝐒𝐝𝐬 —

Offline eriblo

  • Full Member
  • ****
  • Posts: 1334
  • Sweden
  • Liked: 1637
  • Likes Given: 268
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #66 on: 12/07/2023 05:05 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?
How would that help? Once it encounters a problem during entry you blow it up to maximize the potential causalities by ensuring optimal debris spread?

Online Ke8ort

  • Member
  • Posts: 13
  • United States
  • Liked: 16
  • Likes Given: 25
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #67 on: 12/07/2023 05:12 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?
How would that help? Once it encounters a problem during entry you blow it up to maximize the potential causalities by ensuring optimal debris spread?

I think the idea was to do a singular test to collect data on debris spread to show that the risk is low. Obviously blowing up a starship during reentry is going to produce significantly more debris than just letting it burn up. The one issue is that I'm not sure how convincing a test like that would be to the FAA.

Offline eriblo

  • Full Member
  • ****
  • Posts: 1334
  • Sweden
  • Liked: 1637
  • Likes Given: 268
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #68 on: 12/07/2023 05:57 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?
How would that help? Once it encounters a problem during entry you blow it up to maximize the potential causalities by ensuring optimal debris spread?

I think the idea was to do a singular test to collect data on debris spread to show that the risk is low. Obviously blowing up a starship during reentry is going to produce significantly more debris than just letting it burn up. The one issue is that I'm not sure how convincing a test like that would be to the FAA.
I considered if that was what was intended but it seems much easier to get data on a failed reentry by failing a reentry rather than doing something with the FTS...

Online Ke8ort

  • Member
  • Posts: 13
  • United States
  • Liked: 16
  • Likes Given: 25
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #69 on: 12/07/2023 07:29 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?
How would that help? Once it encounters a problem during entry you blow it up to maximize the potential causalities by ensuring optimal debris spread?

I think the idea was to do a singular test to collect data on debris spread to show that the risk is low. Obviously blowing up a starship during reentry is going to produce significantly more debris than just letting it burn up. The one issue is that I'm not sure how convincing a test like that would be to the FAA.
I considered if that was what was intended but it seems much easier to get data on a failed reentry by failing a reentry rather than doing something with the FTS...

Yeah, I agree. Real world data is always better, and I’m sure there will be a chance for spacex to collect debris field data… intentionally or not  ;D

Online TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4324
  • Tampa, FL
  • Liked: 3243
  • Likes Given: 634
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #70 on: 12/07/2023 08:50 pm »
What if a Starship flight termination system had long-life batteries and could be re-armed before reentry?
Use a sacrificial vehicle for a demonstration of that somewhere over the ocean. Then could the FAA be convinced that entry paths crossing inhabited land presented a sufficiently low risk of casualties?

I suspect that they'll have plenty of reentry departure-from-control datasets to curate in short order.  No need to un-safe FTS systems (which sounds like a terrible capability to build into an FTS system).

Online sdsds

  • Senior Member
  • *****
  • Posts: 7106
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 1952
  • Likes Given: 1918
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #71 on: 12/09/2023 09:34 pm »
A blog post by Wayne Hale, notable for its graphical depiction of Shuttle ET 3-sigma debris footprints.
https://waynehale.wordpress.com/2023/12/06/protecting-the-bird-sanctuary/
— 𝐬𝐝𝐒𝐝𝐬 —

Offline Norm38

  • Full Member
  • ****
  • Posts: 1696
  • Liked: 1272
  • Likes Given: 2316
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #72 on: 12/11/2023 04:00 am »
Reading through this thread I was waiting for the FTS discussion.  To me it perfectly illustrates AI conflict theory.
A crash is coming. Swerve towards the school bus or the motorcycle?

For Starship, activate the FTS at first sign of trouble? Try to use what control authority is left to aim?  Aim where?  Activate the FTS or not?  When?
Or do nothing and let fate decide?

The FAA is going to have to consider all of that. Starship EDL will be computer controlled in a way that Shuttle wasn’t.

Online TheRadicalModerate

  • Senior Member
  • *****
  • Posts: 4324
  • Tampa, FL
  • Liked: 3243
  • Likes Given: 634
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #73 on: 12/11/2023 07:17 am »
AFAIK, FTS isn’t part of EDL at all. Once it’s safed at the end of ascent, it’s done.

Offline Twark_Main

  • Senior Member
  • *****
  • Posts: 3577
  • Technically we ALL live in space
  • Liked: 1857
  • Likes Given: 1181
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #74 on: 12/11/2023 07:29 am »
Reading through this thread I was waiting for the FTS discussion.  To me it perfectly illustrates AI conflict theory.
A crash is coming. Swerve towards the school bus or the motorcycle?

For Starship, activate the FTS at first sign of trouble? Try to use what control authority is left to aim?  Aim where?  Activate the FTS or not?  When?
Or do nothing and let fate decide?

The FAA is going to have to consider all of that. Starship EDL will be computer controlled in a way that Shuttle wasn’t.

It's worth remembering that according to Hans the existing Falcon 9 will even avoid individual buildings if it's landing on land. Even if you want to store the GPS location of every building / infrastructure on Earth (and a relative weight score), it's still only a handful of gigabytes. Naturally you'd use a coarse-grained model for high altitude planning and select the appropriate fine-grained tiles as needed, to manage computation cost.



Also, keep in mind this quote that's been on Lars Blackmore's website for almost six years:

Quote
I am particularly interested in chance-constrained optimal planning, that is, finding the best plans such that the probability of failure is below a given threshold.

So yes, exactly that sort of "risk-optimal control" (recalculated live during EDL) seems to be right up his alley....
« Last Edit: 12/11/2023 07:54 am by Twark_Main »
"The search for a universal design which suits all sites, people, and situations is obviously impossible. What is possible is well designed examples of the application of universal principles." ~~ David Holmgren

Offline Twark_Main

  • Senior Member
  • *****
  • Posts: 3577
  • Technically we ALL live in space
  • Liked: 1857
  • Likes Given: 1181
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #75 on: 12/11/2023 08:00 am »
AFAIK, FTS isn’t part of EDL at all. Once it’s safed at the end of ascent, it’s done.

Phew! Good thing Elon's never surprised anyone or defied conventional wisdom before.  8)


No need to un-safe FTS systems (which sounds like a terrible capability to build into an FTS system).

Unless SpaceX plans to have a worker go out to the pad and hit a reset button (hint: they don't), Starship must have the capability to reset and re-arm its FTS autonomously, correct?


Edit: before someone says it, naturally there are countless ways you could have the QD actuate a reset mechanism (pushing a button, shorting two electrical pads, pressurizing a pneumatic tube, etc), but all this really does is ensure that when system fails it does so on the pad, destroying Stage 0. It's actually an anti-safety, risk-maximizing feature! The real solution is to engineer an FTS that only explodes when you tell it to (which you need to do anyway), and boom you're done. :)
« Last Edit: 12/11/2023 08:20 am by Twark_Main »
"The search for a universal design which suits all sites, people, and situations is obviously impossible. What is possible is well designed examples of the application of universal principles." ~~ David Holmgren

Offline eriblo

  • Full Member
  • ****
  • Posts: 1334
  • Sweden
  • Liked: 1637
  • Likes Given: 268
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #76 on: 12/11/2023 02:03 pm »
There is a fundamental difference between ascent and EDL with regard to FTS.

During launch the IIP starts inside an area that has been cleared and is considered safe for the vehicle or debris to fall within. The job of the FTS is to make sure that the IIP does not exit this area. The stack can greatly change its IIP if it goes out of control and a majority of the energy is in dispersible propellant.

During EDL the IIP starts outside the corresponding safe area and the vehicle maneuvers to get it inside. The vehicles ability to change the IIP rapidly diminishes and requires active control. Loss of control is likely to be about as effective at breaking up the vehicle as FTS activation.

The relatively low lift to drag ratio of Starship means that the debris footprint will likely exceed the possible spread in intact IIP. It is also much more feasible to disperse the remaining chemical energy by venting it compared to full main tanks.

Online sdsds

  • Senior Member
  • *****
  • Posts: 7106
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 1952
  • Likes Given: 1918
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #77 on: 12/11/2023 10:08 pm »
What's the "best possible world" scenario for which Starship flight might return safely to its launch site? Consider:

Flight 3: successful belly-flop mid-ocean
Flight 4: successful soft touch-down mid-ocean
Flight 5: successful simulated boost-back and soft touch-down mid-ocean
Flight 6: regulatory approval to overfly Mexico, successful boost-back to Boca Chica and soft catch

Can it realistically get any better than that?
— 𝐬𝐝𝐒𝐝𝐬 —

Offline DanClemmensen

  • Senior Member
  • *****
  • Posts: 5359
  • Earth (currently)
  • Liked: 4196
  • Likes Given: 1694
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #78 on: 12/12/2023 01:14 am »
What's the "best possible world" scenario for which Starship flight might return safely to its launch site? Consider:

Flight 3: successful belly-flop mid-ocean
Flight 4: successful soft touch-down mid-ocean
Flight 5: successful simulated boost-back and soft touch-down mid-ocean
Flight 6: regulatory approval to overfly Mexico, successful boost-back to Boca Chica and soft catch

Can it realistically get any better than that?
Your #4 and #5 can be combined (i.e., #4 skipped). Unlikely but not impossible.

Offline mikelepage

  • Full Member
  • ****
  • Posts: 1205
  • ExodusSpaceSystems.com
  • Perth, Australia
  • Liked: 850
  • Likes Given: 1347
Re: Safe Intantaneous Impact Point Tracks for Starship Recovery
« Reply #79 on: 12/12/2023 02:15 am »
Might be worth highlighting this skip re-entry simulation that OneSpeed did in his "star series" simulations thread. It was done in the context of P2P, but seems relevant here. The key point being that the first skip occurs around 1350km uprange of the final landing site.

One would imagine you could play around with the values, but ISTM that you'd retire most of the risk with the first (most high-energy) skip . Maybe the authorities will be happy that as long as IIP from potential debris created during *that* skip occurs offshore, the rest are considered far less problematic.

So for KSC one could put that first skip in a fairly wide arc in the gulf to cater for a wide range of incoming orbit inclinations, and similarly for Boca Chica this could be a fairly wide arc in the pacific, with the northern end up past the southern tip of Baja.

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1