FAILURE: SpaceX Falcon I Launch III - August 2

[Dalon]

If, however, they are following the prototypical rocket development path of working out the kinks towards the deployment of a highly reliable system, they may end up with a typical rocket industry cost structure in the process.
Which will let you somewhere in between "they will fail" and "they will blow everybody else out of the market".
Which was at least my point here (albeit it obviously has been perceived differently).

You make the assumption that reaching reliability requires a typical rocket industry cost structure.  I don't believe it is a valid assumption.

If there are definite issues that can be recognized and solved, SpaceX will not have to expand into the typical rocket industry cost structure.  They will fix the issues and achieve a reliable vehicle.  Now if their problems are institutional, you may be right, but I've seen no evidence to suggest this.  In fact, I've seen a lot of evidence to suggest exactly the opposite.  All suggestions are that SpaceX has great respect for quality assurance and maximizing redundancy. 

Consider their launches thus far.  Their first launch failed because of a corroded bolt.  The Marshall Island's low elevation and tiny land mass result in a continual salt breeze.  Corrosion is a Massive issue on the atoll.  Corrosion abatement is a continual issue for the military installations there. 

Had they launched from Vandenberg as initially planned, (quashed by their industry competitors) the corrosion issue may never have appeared.  Vandenberg and the Kwaj are at polar opposites of the rust universe.  Still, it's possible, I believe Likely that the first mission would have still have failed due to either the launch 2 or launch 3 issues.  But if not forced to launch from the Kwaj, they'd only be 2 down now, not three.

After launch 1, the corrosion issue was addressed.  To the best of our knowledge it has not reappeared.

The launch 2 anomaly was caused by fuel sloshing.  Multiple fixes were put in place to resolve this.  Belt and suspenders, software and baffles.  We won't know if this was resolved until SpaceX releases its L3 findings.

As yet we've no information as to what caused the failure in launch 3.  If it was fuel sloshing or a corroded bolt, I would be the first to suggest SpaceX has some level of institutional problems that could potentially be solved with a Big Space Methodology. If it has nothing to do with the previous issues, if the exact cause can be determined and fixed, then I would contend that SpaceX is absolutely on track towards their low-cost goal.

As yet, I've seen No evidence to suggest SpaceX has an institutional issue that could be fixed by adopting the typical rocket industry cost structure.  In fact, I think the evidence suggests exactly the opposite.  It suggests that they are working through problems that can only be tested in a live fire environment.  It suggests that once they've experienced each of these show stoppers, they should reach industry levels of reliability.

Of course, only time will tell.

[Analyst]

1) You make the assumption that reaching reliability requires a typical rocket industry cost structure.  I don't believe it is a valid assumption.

2) If there are definite issues that can be recognized and solved, SpaceX will not have to expand into the typical rocket industry cost structure.  They will fix the issues and achieve a reliable vehicle.

3) Now if their problems are institutional, you may be right, but I've seen no evidence to suggest this.  In fact, I've seen a lot of evidence to suggest exactly the opposite.  All suggestions are that SpaceX has great respect for quality assurance and maximizing redundancy. 

4) Consider their launches thus far.  Their first launch failed because of a corroded bolt.

5) The Marshall Island's low elevation and tiny land mass result in a continual salt breeze.  Corrosion is a Massive issue on the atoll.  Corrosion abatement is a continual issue for the military installations there. 

6) Had they launched from Vandenberg as initially planned, (quashed by their industry competitors) ...

7) If it has nothing to do with the previous issues, if the exact cause can be determined and fixed, then I would contend that SpaceX is absolutely on track towards their low-cost goal.

8 ) As yet, I've seen No evidence to suggest SpaceX has an institutional issue that could be fixed by adopting the typical rocket industry cost structure.  In fact, I think the evidence suggests exactly the opposite.  It suggests that they are working through problems that can only be tested in a live fire environment.  It suggests that once they've experienced each of these show stoppers, they should reach industry levels of reliability.

1) Where do you base your beliefs? Three failure in a row say different.

2) Having quality control takes money. Why do you assume other companies waste money with their "typical rocket industry cost structure"? More likely you need this structure and the costs to have a reliable vehicle. You don't simply fix problems. Others will crop up.

3) Sure they have respect, but this isn't enough. Three failures in a row say different. As does launching minutes after an abort.

4) If it has not been the bolt, probably something different would have doomed F1. As you say yourself.

5) It was their decision to move into the middle of nowhere. Being there has one advantage they seem to like for some reason: Secrecy, no pictures. Corrosion is a factor at CCAFS too.

6) The old Titan 4 myth? Who does prevent them from using Vandenberg, CCAFS, WA? Who?

7) So as long as you don't fail twice for the very same reason you are o.k.? You reduce cost by this strategy? You lost me here.

8 ) I have: Three failures in a row. Everything suggests old spaceflight players have a reason to do stuff the way they do. They want profit too and don't waste money for unneccecary procedures. Doesn't mean everything is perfect there. But there must be a reason, don't you think?

Analyst

[dirkthefirst]

I can't see how adopting a traditional cost structure would fix anything. What I think is needed is a little more guidance from somewhere, maybe a few "old hands" onboard to steady the ship and keep things on course for the future.
Quite a few of the mistakes SpaceX have made seem like unnecessary errors**, so maybe a little experience will fix that.

** Not that you can really get a necessary error, I'm not quite sure of the right word to use though.

[pippin]

If, however, they are following the prototypical rocket development path of working out the kinks towards the deployment of a highly reliable system, they may end up with a typical rocket industry cost structure in the process.
Which will let you somewhere in between "they will fail" and "they will blow everybody else out of the market".
Which was at least my point here (albeit it obviously has been perceived differently).

You make the assumption that reaching reliability requires a typical rocket industry cost structure.  I don't believe it is a valid assumption.

Analyst has said it well.

I believe it's a pretty valid assumption since if that assumption did NOT hold true it would automatically mean that all others except SpaceX are a bunch of morons while the evidence shows that these morons do successful launches vs. SpaceX so far have three failures in a row on their books.

I agree that there might be some pork to get by, especially in the case of things developed on cost+ contracts but I am pretty willing to bet that SpaceX original assumption of one order of magnitude will not hold true. And if they end up having half the cost of old.space those companies will carefully look at what SpaceX does differently and adopt what's there in improvements for themselves.

And in the end SpaceX is now mainly running for one of these gov. contracts as well, aren't they?

No, I still don't think they will completely fail, there IS an open spot in the market for privately funded launch vehicles that will be filled but  their  original claim to completely change the rules of spaceflight so far looks pretty hard to meet to me.

[Analyst]

The traditional cost structure is not a goal and won't fix anything by itself, but it stems from adopting industry standards and procedures.

Analyst

[pippin]

I can't see how adopting a traditional cost structure would fix anything.

You don't "adopt a cost structure". You adopt a quality management regime and that lets you end up with a certain cost structure.

[dirkthefirst]

I can't see how adopting a traditional cost structure would fix anything.

You don't "adopt a cost structure". You adopt a quality management regime and that lets you end up with a certain cost structure.

You know perfectly what I mean.

[Steven Pietrobon]

Wow. Three failures in a row. That equals the Soviet N-1 Moon rocket failure record. The Europa I and II also had perfect records with six consecutive failures, followed by cancellation. The hard lessons from Europa were applied to Ariane I which successfully worked on its first flight!

By the way, I noticed a lot of hooting and shouting during first stage flight. I consider it extremely bad karma to do this during the launch. You should only rejoice once the payload is safely in orbit.

Anyways, I'm sure the Space-X team will have much soul searching in the coming days. Hang in there! If you keep trying you will eventually succeed.

[JWE]

Right now this looks like a step back for SpaceX. Is it good to get too far ahead with Falcon 9 when the 'model' has significant flaws? QA is key here. After a long delay, Demo 2 failed in large part due to flawed fuel management software (on both stages)--after the obligatory 'first stage nominal' callouts. Another long delay, greater complexity with a first stage throttle down, and another failure. Great design is nothing without great QA. History is being repeated here. Ullage motors probably won't help, but a more rigorous test/redesign loop probably will. It can't be forced and it will drive up cost.

The quick turn around in the field is something the military wants. Does anyone have more information on military applications for F1?

[William Graham]

Wow. Three failures in a row. That equals the Soviet N-1 Moon rocket failure record.

N-1 had four failures. Falcon is starting to make the Delta III look reliable, though.

[Analyst]

Yup, Delta III reached orbit, twice. Not the planned one, but an orbit.

Analyst

[Jim]

1) You make the assumption that reaching reliability requires a typical rocket industry cost structure.  I don't believe it is a valid assumption.

2) If there are definite issues that can be recognized and solved, SpaceX will not have to expand into the typical rocket industry cost structure.  They will fix the issues and achieve a reliable vehicle.

3) Now if their problems are institutional, you may be right, but I've seen no evidence to suggest this.  In fact, I've seen a lot of evidence to suggest exactly the opposite.  All suggestions are that SpaceX has great respect for quality assurance and maximizing redundancy. 

4) Consider their launches thus far.  Their first launch failed because of a corroded bolt.

5) The Marshall Island's low elevation and tiny land mass result in a continual salt breeze.  Corrosion is a Massive issue on the atoll.  Corrosion abatement is a continual issue for the military installations there. 

6) Had they launched from Vandenberg as initially planned, (quashed by their industry competitors) ...

7) If it has nothing to do with the previous issues, if the exact cause can be determined and fixed, then I would contend that SpaceX is absolutely on track towards their low-cost goal.

8 ) As yet, I've seen No evidence to suggest SpaceX has an institutional issue that could be fixed by adopting the typical rocket industry cost structure.  In fact, I think the evidence suggests exactly the opposite.  It suggests that they are working through problems that can only be tested in a live fire environment.  It suggests that once they've experienced each of these show stoppers, they should reach industry levels of reliability.

1.  No, it is quality rigor drives a cost structure.  Additionally, maintaining a launch team that can handle 40 hour work weeks with vacation, sick time and training adds to it.  Spacex can go balls to the hall and work one F-1 at time but it can't sustain it.

2.  Those issues are institutional

3.  see above.  "Great respect for quality assurance"?  Where do you see?  It took the first flight failure to institute QA buyoffs and close out photos.  See here is traditional cost structure creeping in.  "maximizing redundancy"?  Aside from their spin on the multiple engines in the F9, the rest of spin is  a "simple design" for reliability and not redundancy.

4.  That was not the only issue
5.  That would have happened at VAFB and CCAFS also.  Especially, CCAFS.  One, the material selection was wrong. (bad design).  Two, most launch vehicles enclose the engine section and provide AC to prevent this problem.

6.  That is a plain lie. 
A. The competitors had nothing to do with it
b.  The USAF was justified in not allowing a launch until the Titan was gone
c.  Spacex wasn't ready
d.  Spacex can go back if they want

7.  You contention is wrong.  The fixes are going drive spacex to the traditional structure

8. wrong again.  The problems don't need live test, they are basic problems.  They won't reach "industry levels of reliability" because they eschew the industry processes that enable the level of reliability

[aero313]

The cause of the first abort was due to a turbopump being one percent out of spec, breaching their version of LCC. That is actually a very right LCC, and they may have lauched without issue.

The one percent is irrelevant.  Setting redlines and aborting when you exceed them is standard practice.  Running close enough to the redlines that you regularly exceed them and have to abort is not standard practice.

That's exactly right.  The reason why a responsible launch company develops launch commit criteria is so you do it prior to launch when you have the time to perform the appropriate analyses and tests to validate your criteria.  You specifically DO NOT want to be waiving these criteria at the last minute when your CEO has launch fever.  You may find out that your criteria were overly conservative and thus overly constraining, but you DO NOT make that decision in the last few minutes prior to launch.  The cause of the failure is probably unrelated to the cause of the abort, but the process that led to the waiving of the requirement is fundamentally flawed.

[manlymissileman]

There's been a discussion re: viability of SpaceX's claims of significant launch cost reductions while improving reliability.  For most commercial players or government payloads the payload costs override the launch costs.  Those comm sats and probes cost a lot of money to design and build and operate, often much more money over their lifetime than the launchers with launch services combined.  It's a huge, often one-of-a-kind investment for the customer.  There is a premium placed on reliability, not as much on the lower launch costs.  (Hence insurance costs are on the payload)  They'd rather pay more for a launcher with some kind of track record. 

I could actually reverse SpaceX's model and start at at least current industry level costs/reliability and *then* attempt to bring costs down.  The established launch providers also constantly look to lower costs (why wouldn't they?  they like larger profits too)  I don't know what's better a successful 'expensive' rocket on the first flight or 3 'cheap' failures.  All with live payload.  As it is, even if SpaceX offered a free ride to a commercial entity on their future F9 there wouldn't be many takers until there is a track record of at least their F1 which has a lot of commonality. 

Secondly, even if the next F1 flight is successful it would be great of course and I wish them success, but that'll be just the beginning.  There needs to be a track record of several successful deliveries for an insurance company to begin evaluations, and customers to gain confidence (and then of course the non-gov market is currently mostly in heavier telecom sats which is F9 not F1)  SpaceX needs to demonstrate it can keep up the launch production and operations consistently.

Anyway, I wish spacex luck and perseverance regardless this failure.  They are actually building and launching things unlike most of the altspace projects.  Perhaps some re-evaluations and less aversion to 'conventional wisdom'.  But the competion and insights they could bring in is good for the industry as a whole.

[WHAP]

6) Had they launched from Vandenberg as initially planned, (quashed by their industry competitors) ...

6.  That is a plain lie. 
a. The competitors had nothing to do with it
b.  The USAF was justified in not allowing a launch until the Titan was gone
c.  Spacex wasn't ready
d.  Spacex can go back if they want

SpaceX can go back, but I doubt the Atlas program would want them launching from SLC-3W (which doesn't really exist any more).  IIRC, parts of the first Falcon 1 flight would have landed directly on SLC-3E.  Given their record, lack of an FTS (more than just an engine shut down command, which would increase costs) is a show stopper for them.

[dunderwood]

F1 has FTS.

[Analyst]

Yeh, it terminates every flight before reaching orbit. Sorry, I couldn't resist.

SpaceX can go back, but I doubt the Atlas program would want them launching from SLC-3W (which doesn't really exist any more).  IIRC, parts of the first Falcon 1 flight would have landed directly on SLC-3E.  Given their record, lack of an FTS (more than just an engine shut down command, which would increase costs) is a show stopper for them.

Correct, but it is their (SpaceX) own fault, not the fault of competitors, the range, the government, the big ghost behind the moon or me.

Analyst

[Jim]

F1 has FTS.

It has no destruct system,

[Jim]

SpaceX can go back,

They can go to SLC-4

[dunderwood]

It has no destruct system,

Do you think it needs C4 on it to destruct?  It's a common dome, all you need to do for FTS is vent the fuel tank while leaving the lox tank pressed.