Author Topic: Odds of rocket explosion during launch  (Read 5028 times)

Offline olasek

  • Member
  • Posts: 97
  • Liked: 0
  • Likes Given: 0
Odds of rocket explosion during launch
« on: 01/03/2012 09:51 PM »
I need this info for some work we are doing to re-evaluate dimensions of restricted airspaces around launch sites in the US and its impact on the aircraft traffic.

I would like to know what is an approximate failure rate during an "average" launch - by failure I mean either the rocket explosion or ground triggered destruction of the rocket (rocket veers off course, etc). Also I am only interested in the time frame say 15 sec after it left the pad till about 3 mins. into the flight.

What is your gut feeling this number should be? So far I have been using 1/100. I am primarily interested in US launches (say in the last 25 years) however if you feel you have data for USSR/Russia I will gladly take it for overall comparison. I know that Soyuz rocket performed over 1500 launches so there should a great deal of statistical data.

« Last Edit: 01/03/2012 11:40 PM by Chris Bergin »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32194
  • Cape Canaveral Spaceport
  • Liked: 10839
  • Likes Given: 321
Re: odds of rocket explosion during launch
« Reply #1 on: 01/03/2012 10:06 PM »
I need this info for some work we are doing to re-evaluate dimensions of restricted airspaces around launch sites in the US and its impact on the aircraft traffic.

I would like to know what is an approximate failure rate during an "average" launch - by failure I mean either the rocket explosion or ground triggered destruction of the rocket (rocket veers off course, etc). Also I am only interested in the time frame say 15 sec after it left the pad till about 3 mins. into the flight.

What is your gut feeling this number should be? So far I have been using 1/100. I am primarily interested in US launches (say in the last 25 years) however if you feel you have data for USSR/Russia I will gladly take it for overall comparison. I know that Soyuz rocket performed over 1500 launches so there should a great deal of statistical data.



The FAA should have this data.

Offline olasek

  • Member
  • Posts: 97
  • Liked: 0
  • Likes Given: 0
Re: odds of rocket explosion during launch
« Reply #2 on: 01/03/2012 10:35 PM »
They do however it is not what I want. They have launch failure data but they define failure anytime payload doesn't reach prescribed orbit.
« Last Edit: 01/03/2012 10:47 PM by olasek »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32194
  • Cape Canaveral Spaceport
  • Liked: 10839
  • Likes Given: 321
Re: odds of rocket explosion during launch
« Reply #3 on: 01/03/2012 10:59 PM »
They do however it is not what I want. They have launch failure data but they define failure anytime payload doesn't reach prescribed orbit.

Then you are going to have to find a report on each failure to get that detail

Offline Namechange User

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 7301
  • Liked: 0
  • Likes Given: 0
Re: Odds of rocket explosion during launch
« Reply #4 on: 01/04/2012 12:08 AM »
I don't think you are going to find an answer to your question, at least one that applies generically.

It's a function of so many things unique to that launch vehicle such as design, engineering, processes, technology maturity, etc. 
Enjoying viewing the forum a little better now by filtering certain users.

Offline olasek

  • Member
  • Posts: 97
  • Liked: 0
  • Likes Given: 0
Re: odds of rocket explosion during launch
« Reply #5 on: 01/04/2012 12:38 AM »
Then you are going to have to find a report on each failure to get that detail
Since I am not looking for exact number and only for order of magnitude I was hoping there would be someone here who because of his/her interest in space related topics would provide me with the answer. Like I said, in my opinion 1/100 sounds good to me for now until someone corrects me that I am seriously off.

It's a function of so many things unique to that launch vehicle such as design, engineering, processes, technology maturity, etc. 
I don't think you understood my question. I am only looking for statistical (historical) data. In other words I am looking for this quotient:
(number of blowups)/(number of launches)

« Last Edit: 01/04/2012 01:31 AM by olasek »

Offline Jason1701

  • Senior Member
  • *****
  • Posts: 2238
  • Liked: 70
  • Likes Given: 39
Re: Odds of rocket explosion during launch
« Reply #6 on: 01/04/2012 12:42 AM »
If you don't count the blowups during the first few years of the space age, I'd say your answer is 1/200. The majority of launch failures these days are due to upper stages, not huge explosions after liftoff.

Offline M_Puckett

  • Full Member
  • ****
  • Posts: 477
  • Liked: 91
  • Likes Given: 60
Re: Odds of rocket explosion during launch
« Reply #7 on: 01/04/2012 01:17 AM »
Yep.  I would bet staging failures and engine re-start failures dominate the list.

You generally have the advantage of knowing the status of a main engine before you release the vehicle.
« Last Edit: 01/04/2012 01:48 AM by M_Puckett »

Offline Namechange User

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 7301
  • Liked: 0
  • Likes Given: 0
Re: odds of rocket explosion during launch
« Reply #8 on: 01/04/2012 01:53 AM »
It's a function of so many things unique to that launch vehicle such as design, engineering, processes, technology maturity, etc. 
I don't think you understood my question. I am only looking for statistical (historical) data. In other words I am looking for this quotient:
(number of blowups)/(number of launches)



No, I understood.  If you are trying to do what you claim, then the quotient you state above is meaningless.  It is about more than "just a number".   

One has to understand the meaning and rationale behind that number and the only thing worse than bad data is making a decision based on bad data. 
Enjoying viewing the forum a little better now by filtering certain users.

Offline edkyle99

  • Expert
  • Senior Member
  • *****
  • Posts: 12590
    • Space Launch Report
  • Liked: 3476
  • Likes Given: 700
Re: Odds of rocket explosion during launch
« Reply #9 on: 01/04/2012 02:01 AM »
I need this info for some work we are doing to re-evaluate dimensions of restricted airspaces around launch sites in the US and its impact on the aircraft traffic.

I would like to know what is an approximate failure rate during an "average" launch - by failure I mean either the rocket explosion or ground triggered destruction of the rocket (rocket veers off course, etc). Also I am only interested in the time frame say 15 sec after it left the pad till about 3 mins. into the flight.

Why only three minutes?  A typical orbital launch usually stays below orbital velocity until 9 or more minutes have passed. 

As for the reliability numbers, one quick look would be to consider 2011, when six of 84 orbital attempts world-wide failed.  Of those six, four failed to orbit outright.  One of those four suffered a propulsion failure seven minutes after liftoff, but pieces of the rocket still managed to crash into a house and pepper the landscape around a village.  Another failed about 5.4 minutes after liftoff, but it fell into an unpopulated area.   A third failed probably 4-5 minutes into flight.  The fourth flight suffered a payload fairing jettison failure less than three minutes into flight, but it flew on, gradually falling below its planned altitude and velocity before reentering thousands of miles downrange.

So, as you can see, in 2011 the odds of a rocket faltering and crashing back to Earth, short of orbit, was only 1 in 20.  That wasn't a fluke.  Four of 74 launches in 2010 failed to reach orbit.  Two of those occurred during the first three minutes of flight.  Three of 2009's 78 launch attempts failed to reach orbit.  And so on.  You can find such details among my web pages, as well as among some of the links given on my web site.  Humbly offered:  http://www.spacelaunchreport.com/  (Hint, look on the year-by-year log pages and search for "FTO".)

Keep in mind also that even during a successful launch, pieces of rocket sometimes separate from the vehicle in an unplanned fashion and fall to Earth.  Objects include insulation, ice, and occasionally small pieces of metallic hardware.  One well-known Space Shuttle flight that was successful saw a metallic pin ejected from an SSME during ascent, for example.

 - Ed Kyle
« Last Edit: 01/04/2012 02:49 AM by edkyle99 »

Offline Antares

  • ABO^2
  • Senior Member
  • *****
  • Posts: 5201
  • Done arguing with amateurs
  • Liked: 368
  • Likes Given: 226
Re: Odds of rocket explosion during launch
« Reply #10 on: 01/04/2012 03:53 AM »
1) Ed has the best resource.  I've steered many people to it since I've known about it.

2) The OP stipulates "in the US" and its impact on air traffic.  At some point, US launches would have a fleeting window in which to affect European, trans-Atlantic or African air traffic (or a really unlucky trans-Pacific flight).

3) Active US launchers, and near term, and using Ed's (K+1)/(n+2) where non-orbital = failure:
Atlas V: 0.96 (k=23=n)
Delta IV: 0.93 (k=14=n)
Falcon 9: 0.75 (k=2=n)
Antares: 0.50 (k=0=n)
maybe Delta 2: 0.99 (k=147, n=148)

If you're going to do a projection, use the projected launch rate for these rockets for the year(s) in question.  But, cumulatively, this would have only given a 64% chance that 5 Atlases and 6 Deltas would all have been successful in 2011.  ULA and the USG have processes in place to increase those odds.

It becomes hard to apply statistical methods to a small sample size that's highly process (human not machine) dependent.  And selective statistics become synonymous with bias.
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 olasek

  • Member
  • Posts: 97
  • Liked: 0
  • Likes Given: 0
Re: Odds of rocket explosion during launch
« Reply #11 on: 01/04/2012 05:21 AM »
If you don't count the blowups during the first few years of the space age, I'd say your answer is 1/200. The majority of launch failures these days are due to upper stages, not huge explosions after liftoff.
Thanks Jason and Puckett - this is roughly what I was after.

No, I understood.  If you are trying to do what you claim, then the quotient you state above is meaningless.
No, you did not and please leave it to me what is important or meaningless, you don't even know what the scope of the problem is.

Why only three minutes?
Because after that estimated debris dispersion is so large (very large sigma values in Gaussian distribution) that probability densities fall below certain threshold values. So yes any failures outside this time window (whether of violent or other nature) are of no significance to our work (3 min already has good safety padding, 2 min most likely would be enough)
« Last Edit: 01/04/2012 05:26 AM by olasek »

Offline Namechange User

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 7301
  • Liked: 0
  • Likes Given: 0
Re: Odds of rocket explosion during launch
« Reply #12 on: 01/04/2012 12:15 PM »
No, I understood.  If you are trying to do what you claim, then the quotient you state above is meaningless.
No, you did not and please leave it to me what is important or meaningless, you don't even know what the scope of the problem is.


Then why are you on here asking random people on the internet how to obtain data to inform your solution?

If you truly believe that you can just get a single number, apply it to all launch vehicles that generally have differences in design, maturity, processes, etc then fine.

I'll I was trying to do was point out you have to look deeper than that
Enjoying viewing the forum a little better now by filtering certain users.

Offline khallow

  • Extreme Veteran
  • Full Member
  • ****
  • Posts: 1956
  • Liked: 6
  • Likes Given: 4
Re: Odds of rocket explosion during launch
« Reply #13 on: 01/04/2012 05:19 PM »
As an aside, you probably also ought to look at various restrictions and policies. Consider this: does your spaceport allow first time launches of brand new vehicles? You'll get different failure rates, if say, all of your launches are of new vehicles rather than vehicles with an established flight rate. For example, banning vehicles that don't have three launch successes under their belt will give you a lower failure rate than if you don't. You may find down the road that high failure rate vehicles and prototypes are more likely to be flown at the spaceport.
« Last Edit: 01/04/2012 05:52 PM by khallow »
Karl Hallowell

Offline Namechange User

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 7301
  • Liked: 0
  • Likes Given: 0
Re: Odds of rocket explosion during launch
« Reply #14 on: 01/04/2012 05:59 PM »
As an aside, you probably also ought to look at various restrictions and policies. For example, does your spaceport allow first time launches of brand new vehicles? You'll get different failure rates, if for example, all of your launches are of new vehicles rather than vehicles with an established flight rate. For example, banning vehicles that don't have three launch successes under their belt will give you a lower failure rate than if you don't. You may find down the road that high failure rate vehicles and prototypes are more likely to be flown at the spaceport.


This was, essentially, the point I was trying to make. 

Maturity of systems will likely have an effect.  The root cause of failures also need to be understood.  Was it a temporary spike where an engineering issue or process was at fault, corrected, and no issues since?  Was it a "one of a kind thing" that no amount of engineering and process control could have predicted?  Is it something more systemic (like the recent problems in Russia and what the Russians themselves have spoke about)?

In my opinion a simple "quotient" will hide the real information. 

Enjoying viewing the forum a little better now by filtering certain users.

Tags: