Author Topic: Falcon 9 Range Safety and Potential Abort Modes  (Read 11956 times)

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5481
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 581
  • Likes Given: 679
Falcon 9 Range Safety and Potential Abort Modes
« on: 05/27/2012 07:35 pm »
[Note: this thread was originally titled, "Falcon 9 - Acceptable Risk to Public for Launch." I changed it because ... well, because I could. --sdsds]

----

Now that the launch of Falcon 9 flight 3 (with the C2+ Dragon) is successfully completed there's no risk of jinxing it by discussing range safety. This link:

http://www.gpo.gov/fdsys/pkg/FR-2012-04-24/html/2012-9737.htm

shows the waiver SpaceX requested and received from the FAA regarding the safety of the launch. The juicy bits:
Quote
The Falcon 9 003 launch is the first launch for which the FAA has ever waived the Ec requirement of 0.00003 for launch. The 45th Space Wing Range Safety calculated the collective risk to the public from the Falcon 9 003 launch to be between approximately 0.000098 and 0.000121.

Are the details of range safety calculations available for public review? Neither the FAA nor the AF seem to distinguish between risk to people who knowingly put themselves at risk (i.e. by going to watch the launch) and those who are exposed to risk through no action of their own (i.e. those in Europe at risk from falling debris).

From the launch video it was clear the duration of flight beyond the European gate and before the vehicle reached orbit was short. Was an IIP trace provided for the launch? (Or the presumably quite similar trace of the ground track?)
« Last Edit: 10/01/2012 03:03 am by sdsds »
-- sdsds --

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32634
  • Cape Canaveral Spaceport
  • Liked: 11485
  • Likes Given: 340
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #1 on: 05/27/2012 07:36 pm »
Neither the FAA nor the AF seem to distinguish between risk to people who knowingly put themselves at risk (i.e. by going to watch the launch) and those who are exposed to risk through no action of their own (i.e. those in Europe at risk from falling debris).


Because they are both included.

Offline joek

  • Senior Member
  • *****
  • Posts: 3128
  • Liked: 874
  • Likes Given: 444
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #2 on: 05/27/2012 07:48 pm »
Are the details of range safety calculations available for public review? Neither the FAA nor the AF seem to distinguish between risk to people who knowingly put themselves at risk (i.e. by going to watch the launch) and those who are exposed to risk through no action of their own (i.e. those in Europe at risk from falling debris).

Short version here (and lots-o-links to more reading).

Edit: Sorry, I think you were asking whether specific risk assessments are avaiable to the public?  I've looked in the past and only thing I've found are presentations on methodologies, occassionally with a real-world example and a few numbers (far from an actual assessment).  I'd also be interested if anyone has pointers to more detailed publicly accessible info.
« Last Edit: 05/27/2012 08:18 pm by joek »

Offline Antares

  • ABO^2
  • Senior Member
  • *****
  • Posts: 5202
  • Done arguing with amateurs
  • Liked: 368
  • Likes Given: 226
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #3 on: 05/27/2012 08:19 pm »
I bet it's different because of the payload.  This is the first FAA-licensed capsule launch as opposed to a satellite which is going to be in much smaller pieces if there's a LV RUD while the IIP is over Europe.  Plus, FAA-licensed launches usually go southeast toward GTO or southwest out of Vandenberg to SSO.

The first two Falcon 9 launches went to 35, over sparsely populated Africa as opposed to Europe.
If I like something on NSF, it's probably because I know it to be accurate.  Every once in a while, it's just something I agree with.  Facts generally receive the former.

Offline Silmfeanor

  • Full Member
  • ****
  • Posts: 1199
  • Utrecht, The Netherlands
  • Liked: 275
  • Likes Given: 545
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #4 on: 05/27/2012 09:19 pm »
I bet it's different because of the payload.  This is the first FAA-licensed capsule launch as opposed to a satellite which is going to be in much smaller pieces if there's a LV RUD while the IIP is over Europe.  Plus, FAA-licensed launches usually go southeast toward GTO or southwest out of Vandenberg to SSO.

The first two Falcon 9 launches went to 35, over sparsely populated Africa as opposed to Europe.

Interesting. I was thinking about this. If there was a failure during moment of the launch where the payload would impact europe, let's say a total shutdown of the second stage - do you think dragon would survive the reentry?
The shape should autocorrect assuming everything else burns of during reentry. Things might melt, however - pieces of pica-X will certainly make it to the ground, but how much more? The whole capsule?

Without working or inside info from spaceX, I think it'll be impossible to know - but i'd be interesting if you could possibly load software that detects if something happens and brings the parachute system to some sort of deploy position?
Has this ever been though of on previous missions? I do realize this is the first unmanned craft ( ie without LAS ) that has reentry capability, or poses an issue for reentry.
Perhaps on the very first few manned missions something like this came up before - or the nuclear payloads might be an analogue.

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5481
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 581
  • Likes Given: 679
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #5 on: 05/27/2012 11:04 pm »
Neither the FAA nor the AF seem to distinguish between risk to people who knowingly put themselves at risk (i.e. by going to watch the launch) and those who are exposed to risk through no action of their own (i.e. those in Europe at risk from falling debris).

Because they are both included.

Yes, that makes sense. The FAA waiver provides some insight into which of these dominates the overall risk, referencing a 2005 Titan launch:
Quote
[The Titan IV B-30] risk was mainly attributable to downrange  overflight, as is the case for the Falcon 9 launch.
The reference to risk estimates for day vs. night launch seems (to me) to imply a spectrum of risk to people at the launch site:
Quote
total Ec is currently calculated to be between approximately 0.000098 and 0.000121. The lowest number in the range accounts for a nighttime launch and the upper number accounts for a daytime launch.

The first two Falcon 9 launches went to 35, over sparsely populated Africa as opposed to Europe.

Good point. Regarding overflight of Europe, the narrator on the launch webcast said:
T+8:47: Passing European gate
T+9:07: FTS is safed
T+9:27: Mvac shutdown confirmed
So that seems to imply as long as 40 seconds where the IIP trace was across Europe.

If there was a failure during moment of the launch where the payload would impact europe, let's say a total shutdown of the second stage - do you think dragon would survive the reentry?

SpaceX must have disclosed to the FAA their plan for this contingency. My uninformed guess is that Dragon would detach from its trunk and attempt a soft touch-down under parachutes, leaving the trunk and second stage to disassemble themselves as best they could. ;-)

Short version here

Excellent source of info; thank-you!
-- sdsds --

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32634
  • Cape Canaveral Spaceport
  • Liked: 11485
  • Likes Given: 340
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #6 on: 05/27/2012 11:25 pm »
I would say 20 sec at FTS safed

Online A_M_Swallow

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8687
  • South coast of England
  • Liked: 404
  • Likes Given: 183
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #7 on: 05/28/2012 02:02 pm »
Does the calculation include the probability of collision with the ISS?

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32634
  • Cape Canaveral Spaceport
  • Liked: 11485
  • Likes Given: 340
Re: Falcon 9 - Acceptable Risk to Public for Launch
« Reply #8 on: 05/28/2012 02:48 pm »
Does the calculation include the probability of collision with the ISS?

no, because they don't launch in an orbit that intersects with the ISS

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5481
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 581
  • Likes Given: 679
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #9 on: 10/01/2012 03:39 am »
[Note: this thread was originally titled, "Falcon 9 - Acceptable Risk to Public for Launch." I changed it because ... well, because I could. --sdsds]

I accept the consensus expressed on the SpaceX CRS "SpX-1" mission thread that for this flight there will be no abort mode that avoids loss of the payload. Jim asserts the range safety system would destroy the vehicle to disperse its propellants rather than merely terminating thrust and letting the vehicle fall into the ocean, and that the system that separates the capsule from the trunk would not provide enough distance to keep the capsule safe.

(To be clear I not only accept this, but also support the underlying priorities: tricky abort modes would not decrease the risk to the public. To the contrary, the complexity inherent in their existence might well put the public at additional risk or, at a minimum, make it more difficult to verify that the public would not be at risk in an off-nominal situation.)

Also, I am a big SpaceX fan!

Still I wildly estimate this upcoming flight has only an 86% chance of Dragon reaching its intended orbit, and I'm (perhaps morbidly) interested in what other outcomes might occur. Is there really no possible abort other than a fireball dispersing the Falcon propellants and consuming the Dragon in the process?
-- sdsds --

Offline Lars_J

  • Senior Member
  • *****
  • Posts: 6161
  • California
  • Liked: 665
  • Likes Given: 195
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #10 on: 10/01/2012 03:45 am »
Still I wildly estimate this upcoming flight has only an 86% chance of Dragon reaching its intended orbit, and I'm (perhaps morbidly) interested in what other outcomes might occur. Is there really no possible abort other than a fireball dispersing the Falcon propellants and consuming the Dragon in the process?

What in the world makes you think that after 3 successful flights, the probability of a successful orbital injection is only 86%? It sounds very pessimistic. This is not the 1st flight of the 1.1, this will be a 'vanilla' F9 flight.

I'd be interested to see what your estimations of success was for the previous flights using the same method.
« Last Edit: 10/01/2012 03:47 am by Lars_J »

Offline QuantumG

  • Senior Member
  • *****
  • Posts: 8830
  • Australia
  • Liked: 3832
  • Likes Given: 916
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #11 on: 10/01/2012 03:53 am »
That's about the odds Ziggy always gave to Al too. I blame Don Adams.


Human spaceflight is basically just LARPing now.

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5481
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 581
  • Likes Given: 679
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #12 on: 10/01/2012 04:01 am »
Sorry, I should have admitted up front that my wild estimate used "streak" based thinking that no reputable statistician or probability theorist would accept. In short though, lacking any other knowledge about the underlying distribution, and assuming we are sampling from that distribution at random, I wildly use (n+1)/(n+2) as a predictor, where n is the length of the current streak.

(I think SpaceX has a streak of five successful outcomes in a row, and thus predict with 6/7 confidence that the next outcome will also be a success.)

As for the last COTS flight, obviously it was (5/6)=83%. ;)

Am I a tongue-in-cheek pessimist? Sure. But tell me, do things sometimes go wrong, or not?
-- sdsds --

Offline sdsds

  • Senior Member
  • *****
  • Posts: 5481
  • "With peace and hope for all mankind."
  • Seattle
  • Liked: 581
  • Likes Given: 679
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #13 on: 10/08/2012 02:37 am »
Debris falling from the vehicle at T+1:20 would impact in a zone cleared of all vessels, yes?

Congratulations to SpaceX, BTW. After this (CRS-1) successful launch to orbit, I give their next launch (of any vehicle) a 7/8=88% chance of also reaching orbit. Out of curiosity, now that we've seen what appears to have been a rapid unplanned disassembly of a Merlin in flight, does 88% seem far off?
-- sdsds --

Offline baldusi

  • Senior Member
  • *****
  • Posts: 7442
  • Buenos Aires, Argentina
  • Liked: 1460
  • Likes Given: 4541
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #14 on: 10/11/2012 02:25 pm »
First, it was not a RUD. They stated they kept getting telemetry data from the engine, thus, it was not a RUD. Second, we don't know the root cause. It might have been a materials issue, a process issue or a design issue. It's too early to tell. But if they find the root cause and correct, then there wouldn't be any reason to consider this issue in the next launches.
Since they have flown 38 x (F1+F9) Merlin 1C and 4 Merlin x 1C Vac, and only had one issue, it would give the engine something like a 97% of proven reliability. Next flight will also be a Dragon, for which we might assume that they'll have enough margin to survive another engine out.  Then, the probability of a successful mission if you only consider the engine risk, and you assume there's the same risk for the Vac than the first stage. And further assume that Falcon 9 can survive an engine out (not true at the beginning), but not a second failure (it can later on the mission) and a Vac failure is a LOM. Then you'd have a probability of success of 95.3%.

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 11657
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 8759
  • Likes Given: 7376
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #15 on: 10/26/2013 04:00 am »
Bump.

There are some interesting range safety questions in the latest Grasshopper thread, starting about here, more or less

http://forum.nasaspaceflight.com/index.php?topic=29005.msg906430#msg906430

I am interested in how system design, decision criteria[1], and standard operating procedure needs to change if there are multiple active vehicles.  Also if there would need to be changes in termination to allow a nearby vehicle to "live" if one goes bad.

I am also curious if there have ever BEEN multiple active vehicles that had RSOs watching them. I think not but am probably wrong.

1 - joek said "The RSO's duty is clear and the solution is binary: the vehicle adheres to its allowed flight profile(s), or it does not.  If it does not, it is terminated.  The only question is how termination is accomplished" but what if you go a layer down... what does adherence mean...
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Offline aero

  • Senior Member
  • *****
  • Posts: 3050
  • 92129
  • Liked: 827
  • Likes Given: 294
Re: Falcon 9 Range Safety and Potential Abort Modes
« Reply #16 on: 10/26/2013 05:54 am »
My concern re. termination is that Range Safety terminate the correct stage, not all stages of a Falcon Heavy with RTLS capability. Classically, where a rocket is a single, expendable vehicle, either the vehicle adheres to its allowed flight profile(s), or it does not. With a FH having boosters returning to the launch site, termination needs to address each stage separately because they are separate vehicles.

I see FH flight profiles as:

1. Launch to booster separation, one vehicle treated classically,
2. Boosters separated, giving 3 unique vehicles,
3. Stage 1 Core separated, giving 4 unique vehicles.

It should be clear to everyone that only the vehicle that goes off its allowed flight profile should be terminated. What modifications to the range safety system, if any, will be needed to avoid terminating the remaining healthy vehicles when one (or two or three) goes off profile?
Retired, working interesting problems

Tags: