Before you read this blatant slam piece by @LexNextDC dumping on @SpaceX note that "The Lexington Institute receives funding from many of the nation's leading defense contractors, including Boeing, Lockheed Martin, Raytheon and United Technologies." https://t.co/BtEetuU2Xmpic.twitter.com/jisv7d3aNX
And of course @ulalaunch is (apparently) more than happy for a third party funded by its own parent companies (i.e 2 degrees of separation) to slam @SpaceX like this. Otherwise they'd publicly distance themselves from @LexNextDC – or not.
“In 2017 the latest version of the company’s Merlin rocket engine blew up at a testing facility in Texas.” Isn’t that more-or-less the point of testing? Making sure the explosions are on the test stand rather than the launch pad.
Also saying, in the context of the Zuma launch, that ULA has flown many different payload adaptors is a nice touch. How many of those were provided sight unseen by the customer?
Exactly. The RS-68 had a turbine blade failure in testing too. :shrug:
How many SpaceX payload adaptors have failed, as a matter of interest? None that I recall. Perhaps if SpaceX had been allowed to furnish the payload adaptor there wouldn’t have been a failure? Regardless, the secrecy around the payload would have made interface checks difficult, regardless of the contractor.
If Zuma really was a failure to deploy, I believe that would be a first for Falcon 9 launches. Either with their own payload adapters and deployment systems or customer-provided ones.
Whoever owned/owns Zuma would have to disclose whether it failed or not! Question: Have you stopped beating your wife? Answer “yes” or “no”! Unless we know what the intended mission of Zuma was AND whether or not it achieved at least part of that mission following separation, pointing fingers is useless and prejudicial!
Agreed and I stand corrected! I must have been confusing my decades. I do remember a period where the US, due to several reasons, basically had no satellite launch capabilities. Possibly in the Titan series of the 90’s.
I think you’re referring to Delta III, which had a miserable record. Two major failures in two launches, a partial failure in its third. It never delivered a real payload into orbit. First flight: 1998, last flight: 2000.
After Challenger was lost, between February and May, Titan, Atlas and Delta all had failures – people began speculating about the perennial Russian (Soviet) trawlers typically seen offshore.
Given the reliability history of rockets in general and the cost of failures, one could make a pretty strong case that a rocket with a history of never having had a catastrophic failure could be worth a pretty significant price premium. Aside from Delta IV and Atlas V, the only other rocket with such a claim that comes to my mind is Saturn and it had fewer total missions and a number of incidents that could easily have been catastrophic.
A “pretty significant” premium, yes. But that doesn’t address the question of exactly how significant that premium can be, before the reliability isn’t worth it. If the difference between 90% and 100% reliability is enough money to cover a complete reflight, it wouldn’t make sense to pay for 100% reliability.
Only if you include the cost of a replacement payload in your reflight cost. For a generic model comsat for instance, the payload cost is usually roughly comparable to the launch cost. But what about for something like the Hubble or Webb telescopes? Even worse, what about something with a once every fews years or less launch window, such as most of the deep space probes? Worst of all, what about humans? What is the replacement cost of somebody’s life? To use your number, riding a rocket with a 1 in 10 chance of failure is just not acceptable. Only the most dire wartime necessity can justify that sort of risk taking. It would require multiple orders of magnitude in cost difference to even start thinking about accepting odds like that. Fortunately, I can’t imagine that anyone intends to do so. Quite simply, the Falcon 9 of today has to be proven to be more reliable than its aggregate success ratio since its first flight, which I am sure is what the commercial crew program is working to achieve before certifying it for human flight.
You are talking about the details rather than the underlying idea. At some point, the cost of higher reliability exceeds its value. We can debate whether that is 90%, 95% or whatever, as well as the cost difference. But my point was that there exists a line above which higher reliability is not beneficial.
And you also raise the point that the cost of a failure is not simply financial. That’s true. The cost of a human life is an issue people debate endlessly. The cost of a once-in-a-lifetime scientific opportunity or the cost of a failed mission to people whose careers depend on success is also hard to quantify. Nor is the cost of a high profile failure to national or institutional reputation. But I think it is correct to say that all those costs are finite. Therefore, at some point, the costs of reducing the risk of failure exceed the cost of accepting the risk.
That, in my opinion, means we need to be talking about balancing the costs of a failure and the costs of avoiding a failure. Just saying one launch vehicle is more reliable than another and that the extra cost is worth an unspecified something is insufficient.
Delta II’s few in the 1980’s – almost four decades ago – there was one failure in 1986, There has NEVER been a catastrophic failure of a Delta IV – what is your point?
I’m not sure what’s worse- attacking SpaceX before the failure investigation is complete or maniacally defending them before the failure investigation is complete. Chill out people – we may not know all of the details of the mission or the spacecraft but we will almost certainly have a failure determination – give it some time.
There is an investigation and someday we will know what Zuma is and why it fried.
However, some pundits are jumping to conclusions that the F9 failed, others refrain from sticking their necks out that far…but still use the incident as an excuse to remind everyone the the F9 second stage has exploded twice recently and so imply that to blame the F9 simply follows a convenient pattern for…the pundit…to say that it is par for the course and that we should just use this as the last straw anyway and bag the F9 and go back to the monopoly that charges unnecessary and industry growth crippling amounts of money.
Yes – SpaceX has done a lot to advance launch vehicle technology and reduce costs – and much of it on their own nickel. They have had failures but so have other systems. They may be the cause of a Zuma failure or maybe not – pundits that presume its them at this point given whats publicly known are not being fair.
Maybe, maybe not. It is not clear that the abort system as designed at the time of the CRS-7 failure would have saved a crew if there had been one. If you are referring to the recent Zuma incident referenced in this article and -IF- (a pretty wild guess based on present data) Falcon 9 was the cause, that would probably be a failure to command the spacecraft separation. If that is the case and if it is a common failure mode with the crewed Dragon, then an abort system would probably not save anybody from that failure since the spacecraft didn’t separate. However, there does not seem to be reason to think that there is such a common failure mode at this time, and if there were, there are most likely ample means of mitigating it. The only point being that we must not think of having an abort system as a panacea cure for launch vehicle reliability issues. If there are reliability issues with a launcher, then they must be resolved in order to use that launcher for crewed missions whether there is an abort system or not. Having an abort system is no more valid excuse for accepting reduced launcher reliability than having a parachute is an acceptable substitute for adequate aircraft reliability or wearing a seatbelt is an acceptable substitute for safe driving. It is good to have seatbelts, parachutes, and abort rockets, but they are not excuses for reduced vigilance.
One advantage of flying people is that they can respond to emergencies. In the event of a failure to separate from the upper stage, they would press the button labeled “try again.”
But I don’t see any system level, common mode problems related to the Zuma flight. At least not if it’s a separation issue. That would involve a non-standard payload adapter and a custom interface between it and the second stage. There isn’t much in common with the second stage to Dragon 2 adapter and interface.
But I disagree with your point about seatbelts and parachutes on a more philosophical level. I could be wrong, but don’t test pilots wear parachutes? And rely on them? That’s not neglecting safety, it’s a realistic admission that safety is a result of experience, and experience means mistakes and learning from them. During that learning process, you are depending on safeties and fall-back options like parachutes. At some level, most parents probably feel the same way about seatbelts when their children are learning to drive.
I suppose you could say we shouldn’t put astronauts on a vehicle which is still in that learning process. But NASA seems to think a vehicle is “opperational” after couple flights. Despite being blasted for that unrealistic attitude in the CAIB report.
I agree that it seems a stretch that there would be a common failure mode between whatever happened to Zuma and a commercial crew flight, but then I wasn’t the one making an argument that there was a common mode or that having an abort system would have been protection from that failure. If there is some common mode, then the abort system would not offer protection from it. Only preventing the failure or having a backup separation system could protect from a separation failure. I agree with your assessment of NASA’s own past shortcomings.
Spacex is trying to do too much at once. Ramp up F9 launch rate, fly FHeavy, fly crewed dragon, recover stage 2, design BFR, fly around moon…..not to mention Tesla. But of the 2 commercial resupply programs its done the best. I would have just preferred an incremental growth/development. I hope commercial crew doesn’t end up like the Hubble did when it first launched. http://Www.robert-w-jones.com
This reminds me of sour grapes voiced by my competitors in the energy industry.
The inertia-hobbled, old business school, emergency energy industry business vertical cannot compete easily or effectively with my company’s approach to doing great business:
1) They do not listen to their customers.
2) They believe their customers will stay with them no matter how badly service delivery gets.
3) They believe their own marketing and press of “We are the oldest, we are the best as a result, and you should never consider any other alternatives to our poor and expensive service delivery.”
4) “Our services, hardware, software, etc. you cannot live without, so just give us more money, and we will promise but never deliver on the service delivery you were told to expect, but never received.”
5) “If you decide to go to a competitor, we will give you sloppy service until the termination of the existing agreement between our two companies, because you are now the enemy, a bottom-feeder, and we are going to hurt you for taking food out of our mouths.”
Does this sound familiar? I say let the so-called “establishment” die. They are the specific cause of their own failures, refuse to evolve and improve service delivery, lower costs, innovate, and champion new approaches which assure mission-critical level success.
How many first stage launch vehicles have any of the big players soft-landed, anywhere?
They can’t do it, they are twenty years behind SpaceX, and without a revolutionary product which makes airliner-like reuse of launch vehicle investments obsolete, they will fail.
If modern aviation standards are the goal, I’d say the shuttle was more like aviation in 1920 and Falcon 9 like aviation around 1935. Much farther down the field, but the ball isn’t in that net yet.
STS was as reusable as Falcon 9 is. Are you forgetting the the entire second stage, payload fairing (for non-Dragon missions), payload separation system, and a good portion of the flight control system are not recovered? STS only discarded the external tank and the nozzle extensions on the SRB’s. It would be instructive to compare the cost ratio of recovered versus non-recovered elements for each system, but STS was about as close to reusability as Falcon 9. In theory. In practice, STS implemented reusability on all of its flights. What fraction of Falcon 9 flights have actually implemented reuse so far?
You also need to factor in the number of reflight cycles. The shuttle program did manage to do many cycles, while the Falcon 9 has yet to fly the same first stage more than three times, and (if memory serves) SpaceX doesn’t plan on more that ten cycles.
I’m not taking the relevance of your point here on cycles? Does it really matter how many times the hardware has flown when pricing a single mission except insofar as previous use might reduce the cost?
The demonstrate lifetime of a Falcon 9 first stage is two launches before being retired. If that were the best they could do, resuability would only cut the per launch production costs by a factor of two. I believe they can do better than that, but whether the maximum number of cycles before retirement is five, 10 or 20 makes a big difference to the savings from reuse. (Until it gets to the point where turn around costs dominate over amortized production costs.)
OK, I take your points, as you are largely correct. It’s too easy to focus on the simplicity of the F9 while neglecting the reduced capabilities when comparing to STS- down mass comes to mind as an example.
It’s the simplicity of F9, I think, that I find so damn elegant, particularly when coupled with propulsive landing. Yes it’s true that STS could carry many astronauts, and in some modicum of style, all the while carrying large amounts of down mass.
But was it ever costly! For $1B- the operational cost widely used for each STS mission-you can buy 10 FH missions! That sort of arithmetic really makes folks sit up and take notice. And sure there are lots of variables, making direct comparison difficult.
While this is a one-sided hatchet job, at least it appears to be accurately informed. Many experts (including SpaceX advocates) actually do consider load and go to be a legitimate concern. Will it cause NASA to not fly SpaceX? I doubt it. We do need to find some early footage of Atlas and Delta rockets exploding though. 😉
Cost aside (a big aside, but still) those rockets are among the most reliable in the world and a testament to tens of thousands of engineers and flight operators.
Please re-read Bill’s post, he is not attacking the Atlas and Delta, he is saying that they too had growing pains. Not all mentions of legacy rockets need a reflexive counter attack – footage of early rocket explosions (note – NOT of the current versions!!!) is widely available.
Correct. It’s fun when pundits talk about rocket explosions as if SpaceX invented them.
My point is that a new rocket…any new rocket…needs some *lee-way* when it comes to failures. Especially innovative rockets doing new things.
Here’s a question for the old folks here…how many Atlas and Delta rockets did they smoke before they had a reliable enough vehicle (and policies) to start their current spotless safety record?
Neither Atlas V nor Delta IV have ever had a catastrophic failure to-date, even when they were new. Both were essentially all-new rockets, with very little heritage to the previous Atlas & Delta rockets other than the name. Certainly less heritage than the Falcon 9 had from the Falcon 1 for instance.
You’re thinking of the hardware not the processes and procedures. Atlas V and Delta IV have basically 100% heritage in those respects.
The problem is those processes have a whole lot of dead weight. Every failure results in an investigation and a report with recommendations on processes changes (usually extra work) to prevent a repetition of the problem. Sometimes they can’t definitively identify the cause, and list the likely possibilities. Then identify work to prevent all of them, even though all but one are hypothetical problems. Sometimes they identify potential, hypothetical issue which might be problems and recommend work to avoid them. And many standard practices were added to avoid problems which don’t really exist anymore due to technological changes.
That’s a way to make sure you are doing everything right but at the expense off lots of unnecessary time, effort and money. SpaceX is trying to do what is necessary, but not all the things which no longer make sense. E.g. automating range safety, which reduces ground operations costs and time between launches. But reinventing the process does mean occasional problems and less than perfect reliability.
The Atlas RD-180 engine is derived from the RD-170 used in the Energia and Zenit, while the upper stage uses the Centaur expander-cycle engine that has in one form or another been in srvice for decades. The Delta IV had a failure on the first heavy launch due to a common mode failure of the liquid hydrogen flow sensors in the two booster core stages.
I had a lot more concern with Shuttle during the fueling-first sequence, when the crew and support personnel are around the fueled vehicle for an hour or more with no protection. With load and go by the time they start fueling the vehicle, the crew are buttoned up with the launch abort system active and can be gone in two seconds. The safety of the fueling process is something that SpaceX (and ULA) establish through a substantial number of unmanned launches to ring out the failure modes and correct them in the design.
The underlying problem in safety is that we get wrapped up in interminable debates based on instinctive impressions and speculation. The lesson of C. Edwards Deming was that quality (and, by extension, safety) are quantitative and established by statistical process control, which requires that we execute the process multiple times and establish reliability by testing. With Apollo, Shuttle and Orion, NASA wanted to launch people on a vehicle with virtually no flight experience, and decided that analysis and redundancy could substitute for operational testing. It can’t be done, because most real contingencies are the result of failures we did not anticipate.
That and the shuttle had no launch abort system. Also, didn’t the Soviets have a pad accident where they killed an entire ground crew with a hypergolic fuel leak?
I think you mean the 1960 accident with a prototype ICBM. (A.k.a the Nedelin disaster.) A leak puts it mildly. The second stage accidentally fired during a test, while it was sitting on top of a fueled first stage.
I think we are talking about the same event. 1960 was during the Space Race, the disaster I mentioned took place at Baikonur (both an orbital launch center and an ICBM test center at the time) and, also at the time, orbital launch vehicles and ICBMs were essentially identical.
But the days of the Space Race were also the days of the Cold War. The Soviet Union didn’t say anything about rocket accidents, and essentially all people in the west heard were rumors, dubious statements by defectors and vague information leaked from government sources. For a bit over a quarter century, that way all we knew about the Nedelin disaster. It was frequently described as a failed space launch and the cause was also attributed to various things. The version I gave is based on Russian, post Cold War reports, so there is some hope it is more accurate.
“the failed mission isn’t going to help SpaceX overcome doubts about the reliability of its launch vehicles.”
There’s nothing really untrue about this piece. Except: I read pretty widely about space, and if there’s ‘doubts’ I haven’t seen it.
The rest of it? This is an opinion-based spin and largely true but for emphasis.
Also: did NASA ‘mandate’ HE tank design? I don’t think I’ve seen this characterized as anything but the discovery of an engineering issue, quickly resolved.
Nothing really new to see here.
“In 2017 the latest version of the company’s Merlin rocket engine blew up at a testing facility in Texas.” Isn’t that more-or-less the point of testing? Making sure the explosions are on the test stand rather than the launch pad.
Also saying, in the context of the Zuma launch, that ULA has flown many different payload adaptors is a nice touch. How many of those were provided sight unseen by the customer?
Exactly. The RS-68 had a turbine blade failure in testing too. :shrug:
How many SpaceX payload adaptors have failed, as a matter of interest? None that I recall. Perhaps if SpaceX had been allowed to furnish the payload adaptor there wouldn’t have been a failure? Regardless, the secrecy around the payload would have made interface checks difficult, regardless of the contractor.
If Zuma really was a failure to deploy, I believe that would be a first for Falcon 9 launches. Either with their own payload adapters and deployment systems or customer-provided ones.
Still, this is a huge problem for SX; there are failure cooties floating everywhere.
SX needs to appear – and BE – faultless in this situation.
Whoever owned/owns Zuma would have to disclose whether it failed or not! Question: Have you stopped beating your wife? Answer “yes” or “no”! Unless we know what the intended mission of Zuma was AND whether or not it achieved at least part of that mission following separation, pointing fingers is useless and prejudicial!
The statement is actually untrue – the engine didn’t explode. There was a problem with the plumbing in the test stand, not the engine itself.
Correct. As I recall it was the injection of LOX into a non-operating engine. Oh. Block 5 motor as well!
Nothing new and I notice he didn’t mention all the exploding Delta IVs back in the 80’s pre ULA.
Delta IVs in the 80’s. First flight was 2002.
Delta IV & Atlas V are impressive rockets, for sure, but are they worth the cost?
Agreed and I stand corrected! I must have been confusing my decades. I do remember a period where the US, due to several reasons, basically had no satellite launch capabilities. Possibly in the Titan series of the 90’s.
I think you’re referring to Delta III, which had a miserable record. Two major failures in two launches, a partial failure in its third. It never delivered a real payload into orbit. First flight: 1998, last flight: 2000.
After Challenger was lost, between February and May, Titan, Atlas and Delta all had failures – people began speculating about the perennial Russian (Soviet) trawlers typically seen offshore.
Atlas did not have a failure in 1986. It did have a failure in March 1987, though.
Given the reliability history of rockets in general and the cost of failures, one could make a pretty strong case that a rocket with a history of never having had a catastrophic failure could be worth a pretty significant price premium. Aside from Delta IV and Atlas V, the only other rocket with such a claim that comes to my mind is Saturn and it had fewer total missions and a number of incidents that could easily have been catastrophic.
A “pretty significant” premium, yes. But that doesn’t address the question of exactly how significant that premium can be, before the reliability isn’t worth it. If the difference between 90% and 100% reliability is enough money to cover a complete reflight, it wouldn’t make sense to pay for 100% reliability.
Only if you include the cost of a replacement payload in your reflight cost. For a generic model comsat for instance, the payload cost is usually roughly comparable to the launch cost. But what about for something like the Hubble or Webb telescopes? Even worse, what about something with a once every fews years or less launch window, such as most of the deep space probes? Worst of all, what about humans? What is the replacement cost of somebody’s life? To use your number, riding a rocket with a 1 in 10 chance of failure is just not acceptable. Only the most dire wartime necessity can justify that sort of risk taking. It would require multiple orders of magnitude in cost difference to even start thinking about accepting odds like that. Fortunately, I can’t imagine that anyone intends to do so. Quite simply, the Falcon 9 of today has to be proven to be more reliable than its aggregate success ratio since its first flight, which I am sure is what the commercial crew program is working to achieve before certifying it for human flight.
You are talking about the details rather than the underlying idea. At some point, the cost of higher reliability exceeds its value. We can debate whether that is 90%, 95% or whatever, as well as the cost difference. But my point was that there exists a line above which higher reliability is not beneficial.
And you also raise the point that the cost of a failure is not simply financial. That’s true. The cost of a human life is an issue people debate endlessly. The cost of a once-in-a-lifetime scientific opportunity or the cost of a failed mission to people whose careers depend on success is also hard to quantify. Nor is the cost of a high profile failure to national or institutional reputation. But I think it is correct to say that all those costs are finite. Therefore, at some point, the costs of reducing the risk of failure exceed the cost of accepting the risk.
That, in my opinion, means we need to be talking about balancing the costs of a failure and the costs of avoiding a failure. Just saying one launch vehicle is more reliable than another and that the extra cost is worth an unspecified something is insufficient.
OKB-1 has a winner in Soyuz, don’t forget.
Soyuz has nowhere near an unblemished record.
Delta II’s few in the 1980’s – almost four decades ago – there was one failure in 1986, There has NEVER been a catastrophic failure of a Delta IV – what is your point?
The more you win, the more the establishment will attack you.
Like I always said, if you have Loren Thompspon’s opinion and $4 you can get a cup of coffee.
I’m not sure what’s worse- attacking SpaceX before the failure investigation is complete or maniacally defending them before the failure investigation is complete. Chill out people – we may not know all of the details of the mission or the spacecraft but we will almost certainly have a failure determination – give it some time.
Don’t think I did that, but OK. I’ll reword.
There is an investigation and someday we will know what Zuma is and why it fried.
However, some pundits are jumping to conclusions that the F9 failed, others refrain from sticking their necks out that far…but still use the incident as an excuse to remind everyone the the F9 second stage has exploded twice recently and so imply that to blame the F9 simply follows a convenient pattern for…the pundit…to say that it is par for the course and that we should just use this as the last straw anyway and bag the F9 and go back to the monopoly that charges unnecessary and industry growth crippling amounts of money.
Better?
Yes – SpaceX has done a lot to advance launch vehicle technology and reduce costs – and much of it on their own nickel. They have had failures but so have other systems. They may be the cause of a Zuma failure or maybe not – pundits that presume its them at this point given whats publicly known are not being fair.
There will not be any problems with Crew F9 I guess. But the abort system will save the crew if there is.
Maybe, maybe not. It is not clear that the abort system as designed at the time of the CRS-7 failure would have saved a crew if there had been one. If you are referring to the recent Zuma incident referenced in this article and -IF- (a pretty wild guess based on present data) Falcon 9 was the cause, that would probably be a failure to command the spacecraft separation. If that is the case and if it is a common failure mode with the crewed Dragon, then an abort system would probably not save anybody from that failure since the spacecraft didn’t separate. However, there does not seem to be reason to think that there is such a common failure mode at this time, and if there were, there are most likely ample means of mitigating it. The only point being that we must not think of having an abort system as a panacea cure for launch vehicle reliability issues. If there are reliability issues with a launcher, then they must be resolved in order to use that launcher for crewed missions whether there is an abort system or not. Having an abort system is no more valid excuse for accepting reduced launcher reliability than having a parachute is an acceptable substitute for adequate aircraft reliability or wearing a seatbelt is an acceptable substitute for safe driving. It is good to have seatbelts, parachutes, and abort rockets, but they are not excuses for reduced vigilance.
One advantage of flying people is that they can respond to emergencies. In the event of a failure to separate from the upper stage, they would press the button labeled “try again.”
But I don’t see any system level, common mode problems related to the Zuma flight. At least not if it’s a separation issue. That would involve a non-standard payload adapter and a custom interface between it and the second stage. There isn’t much in common with the second stage to Dragon 2 adapter and interface.
But I disagree with your point about seatbelts and parachutes on a more philosophical level. I could be wrong, but don’t test pilots wear parachutes? And rely on them? That’s not neglecting safety, it’s a realistic admission that safety is a result of experience, and experience means mistakes and learning from them. During that learning process, you are depending on safeties and fall-back options like parachutes. At some level, most parents probably feel the same way about seatbelts when their children are learning to drive.
I suppose you could say we shouldn’t put astronauts on a vehicle which is still in that learning process. But NASA seems to think a vehicle is “opperational” after couple flights. Despite being blasted for that unrealistic attitude in the CAIB report.
I agree that it seems a stretch that there would be a common failure mode between whatever happened to Zuma and a commercial crew flight, but then I wasn’t the one making an argument that there was a common mode or that having an abort system would have been protection from that failure. If there is some common mode, then the abort system would not offer protection from it. Only preventing the failure or having a backup separation system could protect from a separation failure.
I agree with your assessment of NASA’s own past shortcomings.
I read the article before it was noted here, and (unfortunately) did not notice the author, else I would have immediately stopped wasting my time.
Spacex is trying to do too much at once. Ramp up F9 launch rate, fly FHeavy, fly crewed dragon, recover stage 2, design BFR, fly around moon…..not to mention Tesla. But of the 2 commercial resupply programs its done the best. I would have just preferred an incremental growth/development. I hope commercial crew doesn’t end up like the Hubble did when it first launched. http://Www.robert-w-jones.com
This reminds me of sour grapes voiced by my competitors in the energy industry.
The inertia-hobbled, old business school, emergency energy industry business vertical cannot compete easily or effectively with my company’s approach to doing great business:
1) They do not listen to their customers.
2) They believe their customers will stay with them no matter how badly service delivery gets.
3) They believe their own marketing and press of “We are the oldest, we are the best as a result, and you should never consider any other alternatives to our poor and expensive service delivery.”
4) “Our services, hardware, software, etc. you cannot live without, so just give us more money, and we will promise but never deliver on the service delivery you were told to expect, but never received.”
5) “If you decide to go to a competitor, we will give you sloppy service until the termination of the existing agreement between our two companies, because you are now the enemy, a bottom-feeder, and we are going to hurt you for taking food out of our mouths.”
Does this sound familiar? I say let the so-called “establishment” die. They are the specific cause of their own failures, refuse to evolve and improve service delivery, lower costs, innovate, and champion new approaches which assure mission-critical level success.
How many first stage launch vehicles have any of the big players soft-landed, anywhere?
They can’t do it, they are twenty years behind SpaceX, and without a revolutionary product which makes airliner-like reuse of launch vehicle investments obsolete, they will fail.
Let them. It will be good for our society.
Reusable vehicle is the winner and time is money.
Congratulations to the winner.
You mean like the Space Shuttle?
Come on, Ben. STS wasn’t close to reusable.
OK, close. Barely. A very fine piece of American engineering to be sure, with laudable design goals.
But if modern commercial aviation marks the goal post, STS was on the 40 yard line. F9? First and goal…
If modern aviation standards are the goal, I’d say the shuttle was more like aviation in 1920 and Falcon 9 like aviation around 1935. Much farther down the field, but the ball isn’t in that net yet.
STS was as reusable as Falcon 9 is. Are you forgetting the the entire second stage, payload fairing (for non-Dragon missions), payload separation system, and a good portion of the flight control system are not recovered? STS only discarded the external tank and the nozzle extensions on the SRB’s. It would be instructive to compare the cost ratio of recovered versus non-recovered elements for each system, but STS was about as close to reusability as Falcon 9. In theory. In practice, STS implemented reusability on all of its flights. What fraction of Falcon 9 flights have actually implemented reuse so far?
You also need to factor in the number of reflight cycles. The shuttle program did manage to do many cycles, while the Falcon 9 has yet to fly the same first stage more than three times, and (if memory serves) SpaceX doesn’t plan on more that ten cycles.
SpaceX has said ‘at least ten’, and that’s before a major overhaul, not ‘ten and scrapped’.
I’m not taking the relevance of your point here on cycles? Does it really matter how many times the hardware has flown when pricing a single mission except insofar as previous use might reduce the cost?
The demonstrate lifetime of a Falcon 9 first stage is two launches before being retired. If that were the best they could do, resuability would only cut the per launch production costs by a factor of two. I believe they can do better than that, but whether the maximum number of cycles before retirement is five, 10 or 20 makes a big difference to the savings from reuse. (Until it gets to the point where turn around costs dominate over amortized production costs.)
OK, I take your points, as you are largely correct. It’s too easy to focus on the simplicity of the F9 while neglecting the reduced capabilities when comparing to STS- down mass comes to mind as an example.
It’s the simplicity of F9, I think, that I find so damn elegant, particularly when coupled with propulsive landing. Yes it’s true that STS could carry many astronauts, and in some modicum of style, all the while carrying large amounts of down mass.
But was it ever costly! For $1B- the operational cost widely used for each STS mission-you can buy 10 FH missions! That sort of arithmetic really makes folks sit up and take notice. And sure there are lots of variables, making direct comparison difficult.
But 10:1 is very easy to understand.
While this is a one-sided hatchet job, at least it appears to be accurately informed.
Many experts (including SpaceX advocates) actually do consider load and go to be a legitimate concern.
Will it cause NASA to not fly SpaceX? I doubt it.
We do need to find some early footage of Atlas and Delta rockets exploding though. 😉
Good luck finding that footage.
Cost aside (a big aside, but still) those rockets are among the most reliable in the world and a testament to tens of thousands of engineers and flight operators.
And to America.
Please re-read Bill’s post, he is not attacking the Atlas and Delta, he is saying that they too had growing pains. Not all mentions of legacy rockets need a reflexive counter attack – footage of early rocket explosions (note – NOT of the current versions!!!) is widely available.
Correct. It’s fun when pundits talk about rocket explosions as if SpaceX invented them.
My point is that a new rocket…any new rocket…needs some *lee-way* when it comes to failures. Especially innovative rockets doing new things.
Here’s a question for the old folks here…how many Atlas and Delta rockets did they smoke before they had a reliable enough vehicle (and policies) to start their current spotless safety record?
Neither Atlas V nor Delta IV have ever had a catastrophic failure to-date, even when they were new. Both were essentially all-new rockets, with very little heritage to the previous Atlas & Delta rockets other than the name. Certainly less heritage than the Falcon 9 had from the Falcon 1 for instance.
You’re thinking of the hardware not the processes and procedures. Atlas V and Delta IV have basically 100% heritage in those respects.
The problem is those processes have a whole lot of dead weight. Every failure results in an investigation and a report with recommendations on processes changes (usually extra work) to prevent a repetition of the problem. Sometimes they can’t definitively identify the cause, and list the likely possibilities. Then identify work to prevent all of them, even though all but one are hypothetical problems. Sometimes they identify potential, hypothetical issue which might be problems and recommend work to avoid them. And many standard practices were added to avoid problems which don’t really exist anymore due to technological changes.
That’s a way to make sure you are doing everything right but at the expense off lots of unnecessary time, effort and money. SpaceX is trying to do what is necessary, but not all the things which no longer make sense. E.g. automating range safety, which reduces ground operations costs and time between launches. But reinventing the process does mean occasional problems and less than perfect reliability.
The Atlas RD-180 engine is derived from the RD-170 used in the Energia and Zenit, while the upper stage uses the Centaur expander-cycle engine that has in one form or another been in srvice for decades. The Delta IV had a failure on the first heavy launch due to a common mode failure of the liquid hydrogen flow sensors in the two booster core stages.
Yup. I stand convicted of speed reading.
More like careless while commenting
OK! I concede!
https://www.youtube.com/wat…
Here is one.
What about the sequence from _The_Right_Stuff_? It’s examples from around 1960, but I think it included some very early Atlas failures.
I had a lot more concern with Shuttle during the fueling-first sequence, when the crew and support personnel are around the fueled vehicle for an hour or more with no protection. With load and go by the time they start fueling the vehicle, the crew are buttoned up with the launch abort system active and can be gone in two seconds. The safety of the fueling process is something that SpaceX (and ULA) establish through a substantial number of unmanned launches to ring out the failure modes and correct them in the design.
The underlying problem in safety is that we get wrapped up in interminable debates based on instinctive impressions and speculation. The lesson of C. Edwards Deming was that quality (and, by extension, safety) are quantitative and established by statistical process control, which requires that we execute the process multiple times and establish reliability by testing. With Apollo, Shuttle and Orion, NASA wanted to launch people on a vehicle with virtually no flight experience, and decided that analysis and redundancy could substitute for operational testing. It can’t be done, because most real contingencies are the result of failures we did not anticipate.
That and the shuttle had no launch abort system.
Also, didn’t the Soviets have a pad accident where they killed an entire ground crew with a hypergolic fuel leak?
I think you mean the 1960 accident with a prototype ICBM. (A.k.a the Nedelin disaster.) A leak puts it mildly. The second stage accidentally fired during a test, while it was sitting on top of a fueled first stage.
I thought there was something that happened during the Space Race.
I think we are talking about the same event. 1960 was
during the Space Race, the disaster I mentioned took place at Baikonur (both an orbital launch center and an ICBM test center at the time) and, also at the time, orbital launch vehicles and ICBMs were essentially identical.
But the days of the Space Race were also the days of the Cold War. The Soviet Union didn’t say anything about rocket accidents, and essentially all people in the west heard were rumors, dubious statements by defectors and vague information leaked from government sources. For a bit over a quarter century, that way all we knew about the Nedelin disaster. It was frequently described as a failed space launch and the cause was also attributed to various things. The version I gave is based on Russian, post Cold War reports, so there is some hope it is more accurate.
“the failed mission isn’t going to help SpaceX overcome doubts about the reliability of its launch vehicles.”
There’s nothing really untrue about this piece. Except: I read pretty widely about space, and if there’s ‘doubts’ I haven’t seen it.
The rest of it? This is an opinion-based spin and largely true but for emphasis.
Also: did NASA ‘mandate’ HE tank design? I don’t think I’ve seen this characterized as anything but the discovery of an engineering issue, quickly resolved.