Author Topic: SpaceX customers' views on reuse  (Read 319900 times)

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
Re: SpaceX customers' views on reuse
« Reply #340 on: 11/29/2017 03:57 pm »

I'm talking about the actual flight record of F9's. So far that's what 16 flights from last explosion?

That's the disconnect. 

My apologies for not making my PoV clearer. I had thought it obvious from the context of my comments, but obviously not.  I will have to work on making them more comprehensible to you in future.

And your disconnect is that since the current build of Falcon became current, they have 16 of 16 successes.

That's 100% success rate.

To the extent there are known unaddressed issue with the current build, which may be legacy from previous builds, you have a fair point to make.  So, no known point to make.
Long runs of consecutive successes are, of course, required for a launch system to be considered reliable.  They are not, however, indicators or guarantees of a 100% success rate.  R7 once ran up a string of 133 consecutive successes, then did it again a few years later.  F7 still ranks among the world's most reliable launch vehicles.    Nevertheless, R7 failures still occur.

 - Ed Kyle

A difficult to capture statistical feature is the value/weight of present condition(s) versus those in the past when successes or failures were accruing.  As a short term example, the current failure rate of previously flown vehicles is zero, but it is difficult to use that to say whether reflown vehicles will be better or worse than first-flight vehicles which have a longer track record that includes failures.  Long term, discussing R7 in the days of two strings of ~133 successful flights would have not predicted today's failure rate; current conditions are apparently much more powerfully dictating current reliability than those past successes.

Like all statistics, many significant contributing parameters are just too difficult to quantify, so are assumed away (by omission).  Honest statisticians -- is that an oxymoron? -- would add qualifying statement(s) as to what is specifically included and excluded, and what sensitivity the incomplete treatment might have on the conclusions or predictions.

For above examples, the small sample size on reused booster flights would be pointed out as insufficient to make a prediction about reliability of previously-flown boosters vs new.  In the case of R7, the fact that Russia is strapped for cash and high tech imports due to whatever factors, along with the recent R7 failures, would be called out specifically as a potential contributor to today's R7 being potentially much less reliable than the track record would predict.

It still seems that the most committed statisticians are those in the insurance industry (actuaries) who base their business on inclusion of both the easy-to-quantify and the difficult-to-quantify contributors to launch vehicle reliability -- for the next launch.  Price paid for insurance is frequently available to the public.
« Last Edit: 11/29/2017 04:01 pm by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline deruch

  • Senior Member
  • *****
  • Posts: 2422
  • California
  • Liked: 2006
  • Likes Given: 5634
Re: SpaceX customers' views on reuse
« Reply #341 on: 11/29/2017 07:11 pm »
I'm talking about the actual flight record of F9's. So far that's what 16 flights from last explosion?

And because of this F9 might be able to acquire Category 3 certification from NASA. Not sure NRO and USAF have the same standards, but they were looking at collaborating with NASA on this. Bottom line the explosion at this point is probably a non-issue.

Certification is what matters, and we presently don't know what the certification matrix will be for a reusable F9 or a FH. This is what will tell us what future missions might be available and how these customers truly feel about reuse.

EDIT: WAG they are waiting for reusable F9 Block 5 to get Category 3 certification. This could help getting the same certification for FH and BFR easier.

I mean, in theory maybe they could but not in practice.  There have been modifications to the flight configuration that has made all those successful flight.  That's not a big deal, it doesn't make the F9v1.2 Block 4 a different enough vehicle from Block 3 to trigger a whole new certification process.  But it would require that the F9 go through delta-certification and again for Block 4 to Block 5.  So, while it's true that they could have kept going back to delta-cert the vehicle I think they are just waiting to Block 5 and doing it all then.  Though whether they will decide to try to get Cat. 3 certification with even the Block 5 isn't sure.
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Online FutureSpaceTourist

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 48144
  • UK
    • Plan 28
  • Liked: 81623
  • Likes Given: 36931
Re: SpaceX customers' views on reuse
« Reply #342 on: 11/29/2017 07:40 pm »
For the record, CRS-13 booster re-use now official:

https://twitter.com/StephenClark1/status/935910448821669888
Quote
NASA’s Bill Gerstenmaier confirms SpaceX has approved use of previously-flown booster (from June’s CRS-13 cargo launch) for upcoming space station resupply launch set for Dec. 8.

It's probably NASA that gave the approval and the booster was from CRS-11, but here it is.

Online FutureSpaceTourist

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 48144
  • UK
    • Plan 28
  • Liked: 81623
  • Likes Given: 36931
Re: SpaceX customers' views on reuse
« Reply #343 on: 11/29/2017 07:49 pm »
Write-up by Loren Grush:

https://www.theverge.com/2017/11/29/16715910/spacex-falcon-9-rocket-nasa-international-space-station-reusability

Includes:

Quote
Ultimately, NASA will consider flying on used Falcon 9s on a case-by-case basis going forward, the space agency said.

I wonder what NASA’s case-by-case criteria might be? Test/inspection results of the specific booster, cost and/or schedule benefits? All the above?


Offline Negan

  • Full Member
  • ****
  • Posts: 739
  • Southwest
  • Liked: 210
  • Likes Given: 533
Re: SpaceX customers' views on reuse
« Reply #344 on: 11/29/2017 08:21 pm »
So, while it's true that they could have kept going back to delta-cert the vehicle I think they are just waiting to Block 5 and doing it all then.

The possible issue I see with waiting until Block 5 Cat. 3 certification is which 14 flights count toward it. For example what if only the first flight of a block 5 booster counts. Certifying F9 Full Thrust would mean less flights to certify F9 Block 5 and FH. 

Edit: Also I feel because of their Mars plans as well as their interest in NASA's moon plans pursuing Cat. 3 certification makes sense.
« Last Edit: 11/29/2017 08:34 pm by Negan »

Online FutureSpaceTourist

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 48144
  • UK
    • Plan 28
  • Liked: 81623
  • Likes Given: 36931
Re: SpaceX customers' views on reuse
« Reply #345 on: 11/29/2017 09:24 pm »
Perhaps this thread could benefit from a running list of actual data:

Date         Customer Name        Pertinent Event
---------     ----------------          -----------------       

Where a pertinent event might be
* Announcement of plan to accept reused boosters
* Customer comment they will never use reused boosters, or not until X time
* Link to article quoting customers about experience with reuse
* Flights of reused boosters
* Responses to flight successes/failures that affect customer view of reuse

It could be interesting to see a flow of customer decisions over time.

Agreed! Here’s a quick (and incomplete) attempt to get the ball rolling, sourced purely from this thread. So I’m missing things like SES statements prior to first booster re-use and more recent posts on this thread.

But I’m out of time tonight and I think much better to crowd source the info for such a table once there’s a starting point to work with.

Happy to add this to the first post in this thread (and maintain) once we have a reasonable list.

Corrections/additions welcome in this thread or by DM. For speed I’ve limited links to posts on NSF that I could quickly find!

Edit: now added some missing links and earlier SES desire to re-use booster
Final Edit: table now added to first post in thread and extended further

Date (y/m/d)  OrganisationEvent
15/06/17SESM. Halliwell says SES want to re-use F9 booster
16/08/30SES1st booster re-use agreement announced (SES-11)
17/03/15USAFClaire Leon: no plans for reuse, might consider in future
17/03/30SESSES-10 launch: 1st F9 booster re-use
17/03/30SESM Halliwell@post SES-10 launch press conference: 2 of 3 further SES 2017 launches likely to reuse boosters
17/04/11NASADan Hartmann (Deputy Manager ISS Program): Just started discussing reuse; may not be 2017, but shortly after
17/04/06USAFGen Raymond (Head of Space Command) says USAF would be comfortable with reuse
17/05/05BulsatcomAnnouncement that BulgariaSat-1 will launch on a re-used booster
17/06/15IridiumMatt Desch says Iridium would re-use in 2018 (for big discount/schedule improvement)
17/06/23BulsatcomBulgariaSat-1 launch: 2nd F9 booster re-use
17/10/11SESSES-11 launch: 3rd F9 booster re-use
17/10/19IridiumAnnouncement that Iridium NEXT 4 flight will reuse booster
17/11/29NASAGerst makes official long-rumoured booster reuse for CRS-13
« Last Edit: 11/30/2017 11:50 pm by FutureSpaceTourist »

Online john smith 19

  • Senior Member
  • *****
  • Posts: 10350
  • Everyplaceelse
  • Liked: 2430
  • Likes Given: 13606
Re: SpaceX customers' views on reuse
« Reply #346 on: 11/29/2017 09:38 pm »

And your disconnect is that since the current build of Falcon became current, they have 16 of 16 successes.

That's 100% success rate.

To the extent there are known unaddressed issue with the current build, which may be legacy from previous builds, you have a fair point to make.  So, no known point to make.
As is the the 100% rate of Ariane 5 for the last 80+ flights or the Atlas V at 60+

F9 should be better.

So far it is not. 

You keep saying this, but not a single 2nd stage (the ones that actually exploded) have been inspected post flight. Because 2nd stages are not recovered. So there is no opportunity for reuse to directly improve 2nd stage reliability.
exactly the same with both Atlas V and Ariane 5. Yet both have a considerably longer list of successful launches.
Quote from: Norm38
First stages have been recovered, and not a single M1D 1st stage has failed.

So where is the disconnect?  You can't criticize reuse for not improving the reliability of expendable hardware.
That's the interesting point. ULA and Arianspace have no reuse and have achieved very long runs of successful flights. So far SX have only achieved 16. But neither has visibility of the real effects of flight on the second stage.
Quote from: Norm38
You CAN criticize the flight record of F9 versus other launchers, but it's not relevant to a reuse thread.  Except to maybe show that expendable hardware is less reliable than reusable hardware?
Actually give number of flights flown Vs success rate the revers is being demonstrated so far.

That's what I do not understand.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022 TBC. T&C apply. Trust nothing. Run your own #s "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¬cheap SCramjet proposed 1956. First +ve thrust 2004. US R&D spend to date > $10Bn. #deployed designs. Zero.

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
Re: SpaceX customers' views on reuse
« Reply #347 on: 11/29/2017 11:01 pm »

Quote from: Norm38
You CAN criticize the flight record of F9 versus other launchers, but it's not relevant to a reuse thread.  Except to maybe show that expendable hardware is less reliable than reusable hardware?
Actually give number of flights flown Vs success rate the revers is being demonstrated so far.

That's what I do not understand.

The reverse is not being demonstrated.  What is being demonstrated is that new vehicles are more prone to failure than the ones that have established the best track record. (Also true for any vehicles, for that matter -- see R7, Proton.)

In other words, none have better track records than the vehicles that have the best track record -- a silly tautology.

It would be equally illogical to compare Falcon and Proton, and say that this proves reusable vehicles are more reliable.
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline Norm38

  • Full Member
  • ****
  • Posts: 1696
  • Liked: 1272
  • Likes Given: 2317
Re: SpaceX customers' views on reuse
« Reply #348 on: 11/30/2017 12:36 am »
I wasn't comparing F9 hardware against anything but itself. The 1st stage has a better record than the 2nd.  Despite being larger with more engines. Shouldn't the 1st stage have failed by now too? At the least, reuse isn't making 1st stage reliability worse. Maybe that's all we can say.

Bottom line, the F9's failures are fully decoupled from reuse. Where did anyone lead you to believe that the 2nd stage would magically be made more reliable?

Offline speedevil

  • Senior Member
  • *****
  • Posts: 4406
  • Fife
  • Liked: 2762
  • Likes Given: 3369
Re: SpaceX customers' views on reuse
« Reply #349 on: 11/30/2017 12:52 am »
Bottom line, the F9's failures are fully decoupled from reuse. Where did anyone lead you to believe that the 2nd stage would magically be made more reliable?

They are only fully decoupled from reuse if you believe there is 0% likelyhood that there would not have been another (S1) failure, either through explosion or underperformance without either the margins designed in for reuse, or things found during the reuse effort.
This is not clear.
It's also possible they may have learned things that made S2 more reliable, or the additional margins available in S1 have meant S2 does not need to be pushed so hard - making it more reliable and avoiding a flight issue.

For example, if the margins on the engines were 'properly' designed for the payloads on F9B1S1, and not amenable to increasing thrust, S2 would have had to be pushed harder to get performance upgrades.
« Last Edit: 11/30/2017 12:55 am by speedevil »

Offline oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 5304
  • Florida
  • Liked: 5005
  • Likes Given: 1444
Re: SpaceX customers' views on reuse
« Reply #350 on: 11/30/2017 02:59 pm »
This view by NASA will probably sway other customers.
http://spacenews.com/nasa-approves-use-of-previously-flown-booster-on-next-dragon-mission/
Quote
“We get the equivalent reliability of the reused booster that we would expect from a new booster,” he said.(Bill Gerstenmaier, NASA associate administrator for human exploration and operations.)

Offline tdperk

  • Full Member
  • ***
  • Posts: 369
  • Liked: 152
  • Likes Given: 95
Re: SpaceX customers' views on reuse
« Reply #351 on: 11/30/2017 03:14 pm »
As is the the 100% rate of Ariane 5 for the last 80+ flights or the Atlas V at 60+

F9 should be better.

So far it is not.

Something asserted without being demonstrated.

The Falcon 9 family overall has a success a rate of 94.3 percent, 41.5 successes out of 44 intentions to launch.  This includes Amos6 as a failure.

If that rate is maintained, they are "due"  failure any time now.  If they get terribly many more launches without a failure, then by March of 2018 the current build of Falcon will have an excellent success rate, and it is not bad now.

It is superlative now industry wide when seen against the rate of improvement of a key launcher figure of merit, which is the cost to launch.

Offline Semmel

  • Senior Member
  • *****
  • Posts: 2178
  • Germany
  • Liked: 2433
  • Likes Given: 11916
Re: SpaceX customers' views on reuse
« Reply #352 on: 11/30/2017 07:40 pm »
Thanks ed, that is exactly like I understand small number statistics as well. Though I would call it SUCCESS / TOTAL, not SUCCESS / FAILURE.

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
Re: SpaceX customers' views on reuse
« Reply #353 on: 11/30/2017 08:55 pm »
The Falcon 9 family overall has a success a rate of 94.3 percent, 41.5 successes out of 44 intentions to launch.  This includes Amos6 as a failure.
If Amos 6 is included, I come up with the following.  It seems to me that v1.0 was much different than v1.1 and v1.2, so I've grouped the latter two together for some comparisons.  Different engines, different thrust section, etc.  We'll have to revisit the Stage 1 reflight numbers in a year or so.

Variant  Successes   Raw    Point      95%
         /Failures   Rate   Estimate   C/I
----------------------------------------------
v1.0        4/5      0.80    0.71    0.36-0.98
v1.1       14/15     0.93    0.88    0.68-1.00
v1.2       24/25     0.96    0.93    0.79-1.00
----------------------------------------------
v1.1+1.2   38/40     0.95    0.93    0.83-1.00
v1.1+v1.2
 Stg1 Only 40/40     1.00    0.98    0.92-1.00
Stg1 Refly  3/3      1.00    0.80    0.47-1.00
TOTAL      42/45     0.93    0.91    0.81-0.98

Ed Kyle

Again, thanks for a more robust treatment of the statistics.  It is clarifying to note that the 3-for-3 success rate of the reused boosters still yield a much lower confidence interval than the numbers that include failures.

Issues:
1) Your call that there is one v1.0 'failure' is a significant judgement call when the primary payload was delivered, and the secondary waved off due to NASA ground rules.  There was a very high likelihood that the second stage would have delivered the secondary to proper orbit (90+ percent IIRC) in spite of the booster engine failure on ascent.  The ground rule failed, not the rocket.  Calling this entire launch a flat failure is inaccurate at best.
2) AMOS was a test procedure failure that destroyed a rocket and payload.  That's very bad, or even very stupid, but even the insurance companies didn't call that one a launch failure (since it obviously wasn't).
Changing these two cases or their weighting significantly changes the bottom line*.

Bottom line is that statics tell a subjective tale, not (necessarily) an objective one
Those who believe otherwise are naive.

Mark Twain (possibly originally by Disraeli):
Quote
"There are three kinds of lies: lies, damned lies, and statistics."


* The change to the bottom line would be in my subjective judgement a much more accurate representation of launcher reliability -- but still subjective.
« Last Edit: 11/30/2017 09:32 pm by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline meekGee

  • Senior Member
  • *****
  • Posts: 14158
  • N. California
  • Liked: 14046
  • Likes Given: 1392
Re: SpaceX customers' views on reuse
« Reply #354 on: 11/30/2017 10:54 pm »
The Falcon 9 family overall has a success a rate of 94.3 percent, 41.5 successes out of 44 intentions to launch.  This includes Amos6 as a failure.
If Amos 6 is included, I come up with the following.  It seems to me that v1.0 was much different than v1.1 and v1.2, so I've grouped the latter two together for some comparisons.  Different engines, different thrust section, etc.  We'll have to revisit the Stage 1 reflight numbers in a year or so.

Variant  Successes   Raw    Point      95%
         /Failures   Rate   Estimate   C/I
----------------------------------------------
v1.0        4/5      0.80    0.71    0.36-0.98
v1.1       14/15     0.93    0.88    0.68-1.00
v1.2       24/25     0.96    0.93    0.79-1.00
----------------------------------------------
v1.1+1.2   38/40     0.95    0.93    0.83-1.00
v1.1+v1.2
 Stg1 Only 40/40     1.00    0.98    0.92-1.00
Stg1 Refly  3/3      1.00    0.80    0.47-1.00
TOTAL      42/45     0.93    0.91    0.81-0.98

Ed Kyle

Again, thanks for a more robust treatment of the statistics.  It is clarifying to note that the 3-for-3 success rate of the reused boosters still yield a much lower confidence interval than the numbers that include failures.

Issues:
1) Your call that there is one v1.0 'failure' is a significant judgement call when the primary payload was delivered, and the secondary waved off due to NASA ground rules.  There was a very high likelihood that the second stage would have delivered the secondary to proper orbit (90+ percent IIRC) in spite of the booster engine failure on ascent.  The ground rule failed, not the rocket.  Calling this entire launch a flat failure is inaccurate at best.
2) AMOS was a test procedure failure that destroyed a rocket and payload.  That's very bad, or even very stupid, but even the insurance companies didn't call that one a launch failure (since it obviously wasn't).
Changing these two cases or their weighting significantly changes the bottom line*.

Bottom line is that statics tell a subjective tale, not (necessarily) an objective one
Those who believe otherwise are naive.

Mark Twain (possibly originally by Disraeli):
Quote
"There are three kinds of lies: lies, damned lies, and statistics."


* The change to the bottom line would be in my subjective judgement a much more accurate representation of launcher reliability -- but still subjective.
Your view of the the two incidents is accepted by most.

When using statistics as a predictive tool (as is the goal here), we need to apply common sense  as in "how does this incident influence the probability of future events"

Ed's interpretation, which is challenged every time it comes up, does not do that.
ABCD - Always Be Counting Down

Online FutureSpaceTourist

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 48144
  • UK
    • Plan 28
  • Liked: 81623
  • Likes Given: 36931
Re: SpaceX customers' views on reuse
« Reply #355 on: 12/01/2017 12:19 am »
I’ve completed my trawl looking for significant re-use views/events and added the summary to the first post in this thread. Corrections, additions & suggestions welcome.

Offline deruch

  • Senior Member
  • *****
  • Posts: 2422
  • California
  • Liked: 2006
  • Likes Given: 5634
Re: SpaceX customers' views on reuse
« Reply #356 on: 12/01/2017 02:33 am »
I’ve completed my trawl looking for significant re-use views/events and added the summary to the first post in this thread. Corrections, additions & suggestions welcome.

Lol.  You picked basically the worst of all worlds for how to list the date.  I strenuously recommend the ISO format for datesYYYY-MM-DD.

So, instead of 15/06/17 (which is ambiguously possible to read as either June 15, 2017 or June 17, 2015) you use 2015-06-17.  Based on column width in the table it looks like there is enough room to switch to 4 digit years without any issues. This also avoids any contention over whether US (mm/dd/yy) or Euro (dd/mm/yy) dating convention should be used.

Otherwise, it's a great list and I heartily thank you for putting it together (and scdavis for recommending it, great idea)!  I already learned 2 things I had either missed or forgotten earlier. 

My only other suggestion would be to consider using a different color (maybe on the dates since links will remain blue) to highlight either the first actual reuse launch for any given customer OR highlight any given customer's initial statement that they will/plan to reuse.

edit: including highlight suggestion
« Last Edit: 12/01/2017 02:47 am by deruch »
Shouldn't reality posts be in "Advanced concepts"?  --Nomadd

Offline woods170

  • IRAS fan
  • Senior Member
  • *****
  • Posts: 12095
  • IRAS fan
  • The Netherlands
  • Liked: 18197
  • Likes Given: 12158
Re: SpaceX customers' views on reuse
« Reply #357 on: 12/01/2017 06:15 am »
I’ve completed my trawl looking for significant re-use views/events and added the summary to the first post in this thread. Corrections, additions & suggestions welcome.

Lol.  You picked basically the worst of all worlds for how to list the date.  I strenuously recommend the ISO format for dates:  YYYY-MM-DD.

Emphasis mine.
As someone working in the IT industry I strongly endorse this recommendation.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 14158
  • N. California
  • Liked: 14046
  • Likes Given: 1392
Re: SpaceX customers' views on reuse
« Reply #358 on: 12/01/2017 06:26 am »
I’ve completed my trawl looking for significant re-use views/events and added the summary to the first post in this thread. Corrections, additions & suggestions welcome.

Lol.  You picked basically the worst of all worlds for how to list the date.  I strenuously recommend the ISO format for dates:  YYYY-MM-DD.

Emphasis mine.
As someone working in the IT industry I strongly endorse this recommendation.
As someone who used to program , I also endorse this format, but I also advocate for zero indexing the days of the month.
ABCD - Always Be Counting Down

Offline octavo

  • Full Member
  • **
  • Posts: 247
  • Liked: 186
  • Likes Given: 738
Re: SpaceX customers' views on reuse
« Reply #359 on: 12/01/2017 07:03 am »
I’ve completed my trawl looking for significant re-use views/events and added the summary to the first post in this thread. Corrections, additions & suggestions welcome.

Lol.  You picked basically the worst of all worlds for how to list the date.  I strenuously recommend the ISO format for dates:  YYYY-MM-DD.

Emphasis mine.
As someone working in the IT industry I strongly endorse this recommendation.
As someone who used to program , I also endorse this format, but I also advocate for zero indexing the days of the month.

So today is 0 December? Sounds good to me.

Tags:
 

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