Who will compete with SpaceX? The last two and next two years.

[oldAtlas_Eguy]

So I believe that the F9/FH because of it's displacement by bigger and cheaper launchers will never get much above 200/yr.
See graph below:

Who's graph is that? It is interesting wishful thinking, but that is all. For example, I don't think you'll EVER see more FH flights in a year than F9. It ain't happening. Another larger vehicle could overtake F9, but I don't see FH doing that due to the sheer complexity of the setup. But we'll see how smooth the FH introduction is, I could be wrong...

The graph uses the 10X expansion factor for each 7 years for each of the introduced vehicles on their introduction timelines. This timeline would suggest that FH would get to 20 /yr at 7 years from now with F9 reaching a total of 200/yr. As payloads get bigger the ratio of F9 to FH will get smaller and at some point more FH flights than F9 will occur. The unknown is the IITSy and how that will actually affect the FH flight rate. But the ITSy will itself take ~7 years to get to a flight rate of 20. That is an assumption as well as that it will be introduced 7 years from now in 2024. Both of those can be wrong and probably will be. The graph is a graphical representation of my projections of the future to point out that flight rates as discussed earlier look very logarithmic but could at some point become linear and then asymptotic as it reaches some infrastructure or demand constraints. This graphically would look like an s curve where at the beginning the increase would be increasing itself followed by a steady state anual increase (linear) folloed by an anual increase that decreases each year making the flight rate reach an asymptotic max constraint. Where that asymptotic point is is also an assuption. And that assumption is a guess at demand for F9 sized payloads. Where as the flight rate increases the demand for larger and cheaper boosters will increase shifting the launch demand away from F9 toward FH and ITSy. Eventually this will happen to FH and even ITSy as even larger vehicle go into service. But we are getting way out in time from now and a very very speculative timeline events.

[Mader Levap]

- "Anyone" in this case happens to be the COO of SpaceX. You would be well advised NOT to refer to her statements as "fantasy" given that she has extensive experience in both management and "hands-on" work in the aerospace industry.

It wouldn't be first or last time when someone in this kind of position said something unrealistic (as Dilbert's boss said, optimism is not a crime). Or simply misinterpreted. So excuse me if I won't be that impressed.

- "Not any time soon" is actually first half of next year (which in the rocket business is soon).

I like how some people are flatly ignoring long, long history of SpaceX being late to absolutely everything and pretending it is not case anymore.

It won't be "first half of next year", period. In any sense and any interpretation of "24h turnaround".

- "Many people interpreted it as..." is your way of talking around the issue. 24 Hour turnaround clearly refers to the same stage. Only you see that differently.

Nope. Even in this very thread already are comments (not mine) claiming you can interpret her words differently. As counting only certain parts of process. Or as counting work hours done on stage, not physical days.

I will believe when I will see it. And it is possible I will, just not in timeframe SpaceX says it will be - but way, way, wayyyyy later (if it happens at all).

Hell, it will take decades before there will be so many payloads to launch that 24h rotation (as in actually launching rocket once per day from same pad) is desired, let alone required. And even then only 24h period between launches is needed, not between stages.

I stand by my words. Right now and for at least two decades 24h turnaround is not even needed. At most, you could see forays into making quick refurb or something like that, but actually physically relaunching same rocket within 24h is simply fantasy in this time range and probably even longer.

[Lar]

Mader,  while you don't have to believe stuff, your tone isn't helpful in this sort of discussion. Basically, you were proven wrong about the 24 hour thing (it was pointed out clearly who said it and what was meant) and instead of saying "oops, thanks for the correction, I was wrong about what was meant but I still don't believe it"  you just got more belligerent.

Don't do that, we're all peers and there is no need to be so combative. Thank you. (this post was in lieu of just deleting your post outright)

[jpo234]

I like how some people are flatly ignoring long, long history of SpaceX being late to absolutely everything and pretending it is not case anymore.

This time it might indeed be different. The Block 5 changes are needed for Commercial Crew (upgraded COPVs and turbo pumps), so there is good reason to think that they will be ready soon.

[Brovane]

- "Not any time soon" is actually first half of next year (which in the rocket business is soon).

I like how some people are flatly ignoring long, long history of SpaceX being late to absolutely everything and pretending it is not case anymore.

What is always late is Elon Musk's (reality distortion) time frames.

When Shotwell says we can expect the Block-5 Falcon-9 to fly in early 2018, I expect that will happen. 

If the F-9 Block-5 flies in the first half of next year are you still going to claim it is late anyway because SpaceX is late on Commercial crew? 

[meekGee]

Other than the COPV use and placement, what uncompensated failures has SpaceX had since it's first F1 success?

The better question is, "what unidentified failure modes has SpaceX not yet experienced?" in its 41 Falcon 9 launch campaigns (only 21 by the current variant).  Space Shuttle flew, what, 112 times before the STS-107 failure mode occurred? 

The same question can be asked of any current launch vehicle.

 - Ed Kyle

Ed - extensive wing damage from falling debris occurred numerous times before STS-107 - it's a really bad (and painful) analogy.

Your question can be applied universally, to any launch vehicle, at any time.

Of course hidden daemons may still lurk, but the more you fly, and especially with a reusable rocket, the more confidence you gain.

I think it is a potentially-apt analogy because the true seriousness of the falling foam hazard was not appreciated before the launch.  It wasn't even appreciated during the flight.  SpaceX had issues with He pressurization systems before its first FTO, yet that turned out to only be the first of two launch vehicle losses caused by the pressurization system.

The trigger for this discussion was a disagreement about failure rate for Falcon 9.  Once claim was 1-in-20 (0.05).  A rebuttal claimed 1.5-in-40 (0.0375).  I'll note that v1.2 is 20 for 20 in launches and 20 for 21 in campaigns.  My assessment is that v1.2's failure rate can be confidently said to be at least 1 in 11 or better given current launch numbers and results.  That's comparable to its peers at the same point in their programs. 

The same cannot be said for all Falcon 9's taken together as a family.  Their combined failure rate trails that of Atlas 5 and Ariane 5 and is only slightly better than Titan 4 and Proton.

 - Ed Kyle

With STS, it was a recurring issue, that after Atlantis landed (in 1988) was recognized as having avoided LOC by nothing more than luck, and then promptly ignored.   Completely different than saying "SpaceX may have un-noticed daemons lurking like STS had".

Anyway -

I think that saying "SpaceX had issues with He pressurization systems before its first FTO, yet that turned out to only be the first of two launch vehicle losses caused by the pressurization system." is mis-characterizing it.

Yes, both were Helium-related, but hardly similar.  One was a structural bracket.  The other, technically, not even an He issue but LOX combusting with Epoxy under conditions related to He loading.

I agree with 0/20 and 1/21 respectively, but I will argue that extrapolating these numbers (in either flavor) is nonsensical.  These are not statistical issues.  This was a design issue that was fixed. It doesn't mean that there are others, or that there aren't.

Sometimes, there are issues that indicate poor culture, for example. Those are statistically significant.  Someone forcing a sensor to be installed backwards in a good example.  This wasn't the case though.

When we don't have data, using "whatever data we do have" is often worse than just saying "we don't have data".

If I was an insurance company, post Amos, I'd say that I estimate the odds of F9 having an accident are exactly the same as the odds I estimated pre-Amos.

[envy887]

Other than the COPV use and placement, what uncompensated failures has SpaceX had since it's first F1 success?

The better question is, "what unidentified failure modes has SpaceX not yet experienced?" in its 41 Falcon 9 launch campaigns (only 21 by the current variant).  Space Shuttle flew, what, 112 times before the STS-107 failure mode occurred? 

The same question can be asked of any current launch vehicle.

 - Ed Kyle

Ed - extensive wing damage from falling debris occurred numerous times before STS-107 - it's a really bad (and painful) analogy.

Your question can be applied universally, to any launch vehicle, at any time.

Of course hidden daemons may still lurk, but the more you fly, and especially with a reusable rocket, the more confidence you gain.

I think it is a potentially-apt analogy because the true seriousness of the falling foam hazard was not appreciated before the launch.  It wasn't even appreciated during the flight.  SpaceX had issues with He pressurization systems before its first FTO, yet that turned out to only be the first of two launch vehicle losses caused by the pressurization system.

The trigger for this discussion was a disagreement about failure rate for Falcon 9.  Once claim was 1-in-20 (0.05).  A rebuttal claimed 1.5-in-40 (0.0375).  I'll note that v1.2 is 20 for 20 in launches and 20 for 21 in campaigns.  My assessment is that v1.2's failure rate can be confidently said to be at least 1 in 11 or better given current launch numbers and results.  That's comparable to its peers at the same point in their programs. 

The same cannot be said for all Falcon 9's taken together as a family.  Their combined failure rate trails that of Atlas 5 and Ariane 5 and is only slightly better than Titan 4 and Proton.

 - Ed Kyle

Falcon 9 as a family is comparable to Ariane 5 at the same point in it's program. It took Ariane over 12 years to conduct 41 A5 missions, and 4 were failures or partial failures. F9 took under 7 years and had 3 failures or partial failures to get to 41 missions.

After that, A5 has had 53 straight successes and counting, and it's not at all implausible that F9 can do the same.

[Jim]

I think that saying "SpaceX had issues with He pressurization systems before its first FTO, yet that turned out to only be the first of two launch vehicle losses caused by the pressurization system." is mis-characterizing it.

No, it shows that there were systemic issues and they were lucky that they only lost two vehicles.

[meekGee]

That's just an unsupported statement.

Using big words doesn't make the argument more convincing - what's systemic about it?  Based on what are you saying that they are lucky they didn't have more failures?  What makes that statement inapplicable to any launch vehicle out there?

This is no different then Ed's assertion that there are probably more unrecognized daemons in the design, because look, STS.

[Kansan52]

Are the systemic issues solved, being solved, or ignored?

[Brovane]

Are the systemic issues solved, being solved, or ignored?

Well, SpaceX isn't going to share that except to say they are being dealt with. 

SpaceX made several design choices to maximise payload for the F9.  This includes putting the He tanks inside the LOX tank and using COPV's.  Only long term launch success/failures will determine if they have been solved. 

SpaceX is on an engineering frontier to lower the cost of access to space.  This means that SpaceX will design their rockets differently than what has done before.  Also SpaceX does sparse Matrix engineering and doesn't track down everything.  This means that mistakes will be made. 

[Kansan52]

Thanks.

[tdperk]

I like how some people are flatly ignoring long, long history of SpaceX being late to absolutely everything and pretending it is not case anymore.

Late to be first to have an economically re-usable rocket.  Late to be first to substantively drop the cost of space access.  It's a lot of good "late" for some other company to compete with.

[oldAtlas_Eguy]

This whole item is the fallacy of statistics. The application of a general case that is a statistical event over a large population (the population here being many many differently designed complex systems not the multiple flights of a single system) to specify how a single system will behave in the future is the fallacy. An unknown-unknown is by definition not statistically definable as an analyzable risk. The unknown is that there are from the statistical population a very wide range to what is possible from no matter how many you try none fail to almost every one you try fail from something different.

[Robotbeat]

Whatever. Ed's statistic are fine, although I quibble about how he defines failure.

We use these crude methods because we don't /know/ anything with any kind of quantitative certainty except what the failures and successes tell us. How do you /quantify/ differences in culture, procedures, whether "demons" are being taken care of or not?

You can't. Especially not us. So use statistics. Works crudely with a small sample, it has large error bars in such a case, but it still works.

SpaceX is now launching quickly, on pace for like 18 launches per year. By the end of next year, if they avoid failure, they'll have 40-50 consecutive successes. That's top of the class reliability, on par with Ariane 5, etc. So we don't have long for SpaceX to prove their reliability. And if they suffer failures? Well, then we know that's how things are. Either way, a bigger sample size and more insight.

But SpaceX won't fly Falcon 9 forever. ITSy doesn't have COPVs for pressurization. SpaceX has the opportunity to address any systemic problems with Falcon 9.

[oldAtlas_Eguy]

Whatever. Ed's statistic are fine, although I quibble about how he defines failure.

We use these crude methods because we don't /know/ anything with any kind of quantitative certainty except what the failures and successes tell us. How do you /quantify/ differences in culture, procedures, whether "demons" are being taken care of or not?

You can't. Especially not us. So use statistics. Works crudely with a small sample, it has large error bars in such a case, but it still works.

SpaceX is now launching quickly, on pace for like 18 launches per year. By the end of next year, if they avoid failure, they'll have 40-50 consecutive successes. That's top of the class reliability, on par with Ariane 5, etc. So we don't have long for SpaceX to prove their reliability. And if they suffer failures? Well, then we know that's how things are. Either way, a bigger sample size and more insight.

But SpaceX won't fly Falcon 9 forever. ITSy doesn't have COPVs for pressurization. SpaceX has the opportunity to address any systemic problems with Falcon 9.

And introduce new ones. That is the problem. New systems have new failure modes. That is where the philosophy of engineering for failure tolerance becomes significant. The F9 first stage is failure tolerant to a single and some case dual engine failures. But in other areas it is not. Make the engineering choices where you can get the most for the least costs.

[Robotbeat]

So you fly it a few dozen times launching mass produced satellites before putting people on it, addressing any issues as they crop up. Problem solved.

[oldAtlas_Eguy]

So you fly it a few dozen times launching mass produced satellites before putting people on it, addressing any issues as they crop up. Problem solved.

Definitly one solution which SpaceX is currently doing not only for the F9 but the Dragon as well. That is even though the D2 has a lot of new hardware it also shares many things that have shown reliable operation.

[AncientU]

This whole item is the fallacy of statistics. The application of a general case that is a statistical event over a large population (the population here being many many differently designed complex systems not the multiple flights of a single system) to specify how a single system will behave in the future is the fallacy. An unknown-unknown is by definition not statistically definable as an analyzable risk. The unknown is that there are from the statistical population a very wide range to what is possible from no matter how many you try none fail to almost every one you try fail from something different.

Well put!

Whatever. Ed's statistic are fine, although I quibble about how he defines failure.

We use these crude methods because we don't /know/ anything with any kind of quantitative certainty except what the failures and successes tell us. How do you /quantify/ differences in culture, procedures, whether "demons" are being taken care of or not?

You can't. Especially not us. So use statistics. Works crudely with a small sample, it has large error bars in such a case, but it still works.

...

The major fallacy is thinking these crude methods can be accurate to a few significant digits.   

The second major fallacy is assuming away factors that are difficult to quantify:
1. How likely is the fault to be repeated?
2. How systemic is(are) the fault(s)?
3. How much does culture adjust the statistics?
4. How much does rapid development (or stasis) affect the numbers?
5. How does deterioration in suppliers' quality system affect the likelihood of failure?
6. How do near misses get counted? (As successes? As partial failures? As failures?)
7. How does analyzing returned cores improve vehicle reliability?
8. How much does a demoralized, overworked, down-sizing workforce affect reliability?
9. Add your items here...

These are tough-to-quantify factors, but assuming that they are not in play makes the crude methods even cruder.  Think about all of the 'factors' that a full post-accident review board would list as contributing...

Engineering is done with numbers is not equivalent to having a number is engineering.

[JamesH65]

I like how some people are flatly ignoring long, long history of SpaceX being late to absolutely everything and pretending it is not case anymore.

Late to be first to have an economically re-usable rocket.  Late to be first to substantively drop the cost of space access.  It's a lot of good "late" for some other company to compete with.

It's a bit like saying Usain Bolt was late in the 100m Olympic final because he didn't break the world record.

Of course, he still won. By a long way.


It's not how quickly he got to the tape. The important bit is that he got there before everyone else.