Watching SpaceX Compress The Rocketship Design Process
IMHO @JimBridenstine @ElonMusk It is Interesting to watch @SpaceX compress design iterations of launch vehicles to months or a year while it takes @NASA decades to do so. "SpaceX goes all-in on steel Starship, scraps expensive carbon fiber BFR tooling" https://t.co/49R9boXzc9 pic.twitter.com/MGU8IqWYwc
— NASA Watch (@NASAWatch) March 21, 2019
This cries for double standard. Had NASA done the same thing there would be an outcry for wasting taxpayer’s money.
Maybe a lengthier design process would have uncovered whatever problems with the carbon fiber approach were discovered later on.
But nice way to spin the story. Just sayin.
Actually a lot of people are critical of NASA not changing direction when it’s needed. Often the sunk cost fallacy rules at NASA.
As far as I can tell, those people who are critical of NASA (over that sunk cost fallacy) aren’t in congress. NASA’s likely to pay far more attention to the people who sign the checks.
Yes, one of the very worst fears a government bureaucrat has is going before a Congressional Committee and saying, “oops, we made a mistake, we need to junk everything and start over”. They would rather continue digging the hole deeper, hence SLS/Orion/Gateway.
Agreed. I’m sure some of those bureaucrats are just “running out the clock” until they retire (maybe to a cushy contractor job, after the legally required waiting period, of course).
Would you agree that this is one of the chief factors that separate governmental efforts and private enterprise efforts? This “fear of making a mistake”, as you aptly phrase it, runs quite deep.
Humility, too, can be difficult to find.
Lest anyone think I am painting with a broad brush: it’s the policy wonks that I’m describing here.
NASA could be more proactive about telling Congress what it needs and what it doesn’t need. I had been told since SLS was created that it would be “political suicide” for a NASA Administrator to say anything negative about SLS. I disagreed with that assertion and believe that a NASA Administrator ought to be more honest and forthcoming with Congress. NASA never needed SLS as it’s a dead-end project. Too high cost and too low flight-rate from the very beginning.
But, since we’re actually seeing that happen today with NASA Administrator Jim Bridenstine, I guess we’ll see what happens. Hopefully this sets a positive precedent for the future.
Which leads to the conclusion that the Administrator and the Senator had a frank talk before any of this came out; I’d love to have heard that conversation.
Which is why innovation has historically come from private industry and not government. The space industry was the exception where government took over the role of innovator during Project Apollo and its why it has been stagnate for decades, as the SLS/Orion demonstrate so well.
There’s an awful lot of research and development with the NASA imprint, Dr. M.
The main criticisms of NASA are not for making changes early in the development process but for starting projects then cancelling them after several years and huge amounts of money. Or having forced cancellations of large projects when they are hopelessly over budget and behind schedule. That’s a waste not only of any hardware and tooling that may have to be discarded but even more of a problem is the massive amounts of man-hours that end up being totally wasted, after subtracting whatever research might be applicable to future projects.
I think everyone would be thrilled if NASA only wasted as much time and expense as SpaceX did on carbon-fiber. Also they moved away from carbon-fiber quite a while ago but only recently decided to destroy the tooling.
Yes it’s the other side of the coin. From NERVA to X33, to inflatables and other examples.
Yep, a very, very long time, maybe two or three months ago…
At the pace that SpaceX moves that is a long time ago.
“Had NASA done the same thing there would be an outcry for wasting taxpayer’s money.”: Nope, that’s not what happened. NASA tried to do exactly the same thing in X-33, when the composite tank failed there is a proposal on the table to replace it with metal tank, today pretty much everybody wished they have gone with that route and at least get X-33 to fly.
If SpaceX added a dozen or two committees, they could slow down to the level of the Big Boys…
An elephant: That’s a mouse built to government specs!
And a long over-due shaking up it is. Watching the old-schoolers response is also very entertaining
“WTF!?! THEY can’t do THAT!!”
“Dude, they just…did” ?
“It is Interesting to watch @SpaceX compress design iterations of launch vehicles to months or a year while it takes @NASA decades to do so.”
When people in an organization are free to focus their attention on engineering, rather than what will please “the Committee of 535” in Congress, a real lot of engineering can get done in a few months. May that continue. May it spread and grow!
“When a society moves from allocating resources by custom and tradition (moderns read here, by politics) to allocating resources by markets, they may be said to have undergone an industrial revolution” Arnold Toynbee-1884
SpaceX is fortunate. They do not have a senator from Al telling them how to design a rocket.
No one deals with sunk costs better than Musk.
Why pay for rent on the lot plus maintenance on the “tent” and etc. to store a piece of tooling they’re never going to use? I think they did the right thing here.
Fail early, rather than later, in order to minimize the long term costs.
Yes, unlike NASA they are willing to admit they made a mistake in design, write if off, and move on, while NASA just keeps doubling down on the SLS/Orion/Gateway…
Meanwhile reports are Boeing has moved their first demostration of the CST-100 back several months again.
I don’t thin that ‘willing’ is the correct verb here; more likely ‘able’.
I’d disagree with the “admit they made a mistake” part instead. That may not be the obstacle. I think an important issue is how senior management within NASA, or Presidents and Congressmen, will react. I think people are _able_ to have a very unpleasant conversation with their managers, or get grilled in front of a congressional committee, and explain and justify their decisions. If they are truly the right people to be making those decisions, they ought to be able to defend them. But that would be about as fun as dental surgery and potentially bad for their careers. That can make people _unwilling_ to get into those sorts of situations.
Remember, first priority of NASA is not space, its JOBS! SpaceX has laid off a significant number in their design shift. NASA doesn’t have that luxury.
Doesn’t Boeing use similar fiber winding composites for their 787 fuselage.
As for design cycle times … smaller companies have shorter cycles than larger ones. Large companies have more inertia than small ones. Fact of life.
That fuselage is not subject to the same level of stress and heat, though.
The 787 fuselage is a hair under 6 meters wide. The Starship vehicle SpaceX is trying to build would have a 9 meter diameter. My understanding is that making a composite structure 50% larger is way more than 50% more difficult.
If I recall, in the article it mentioned that while the booster is currently an all metal design, the starship is planned to have composite structures. It wasn’t clear from the photo that the tooling actually was totally destroyed. The tent yes.
So, the essence of their superiority is that SpaceX is able to scrap a design in months, while it takes NASA years? 🙂
SpaceX is following a pretty sensible process that I’ve usually seen with – of all people – the Russians (*groan*), on a lot of their rockets, planes, ships, tanks…. Essentially, it boils down to:
— Accept the “unacceptable” idea of Added Weight,
— Simplify & Scale-up and
— Spend more on Propellant to lift it all
(good grief, the stuff they’ll use is not some exotic bank-busting dilithium crystals).
I look at this Starship beastie in the above photo and compare it to a Brawny Mac Truck, versus NASA’s attempt to build a carbon-fiber, hybrid, Super-Car (very fast, energy-efficient, uber-safe and sexy…. but you’ll need to sell your house ten times over, and wait as many years, just to get something that’ll likely break down. )
I guess that’s what makes Musk unique from the rest of us to give the call “scrap it!” of a major development effort and then focus on a new effort. Many others have done the same, I’m thinking about projects where a manager does similar but creates anguish (and anger) from various sub-contractors that get burned from lost work.
I suppose that, implicit in a story of this kind, there’s a natural comparison between NASA and SX. But be careful here. It’s hardly a fair comparison.
Why? Different arenas, that’s why. Completely different playgrounds, so to speak. Different strategies, and more important, different rewards. NASA has hoards nipping at their heels, each with a smug sense that as a citizen they can do a better job.
Similarly, NASA has lots of web sites…well, like this one, actually…that legitimately find ‘news’ in places that simply don’t exist for SX, where failure is heralded as a way to learn and move forward, while NASA failure is a waste of taxpayer money, or more pork, or whatever. And they are right, too. But at the same time realize it’s not exactly the same thing.
This is one reason why it is dangerous to compare ‘private enterprise’ vs ‘government’, a comparison that sometimes works at the largest scales, but down in the dirt isn’t helpful.
Michael, you are bemoaning the fact that people celebrate the advantages of the freedom that private companies, without the influence of politically allocated funding have, over those bound to the demands of politicians to help get them re-elected.
“This is one reason why it is dangerous to compare ‘private enterprise’vs ‘government’, a comparison that sometimes works at the largest scales, but down in the dirt isn’t helpful.”
Isn’t helpful to who???
Is dangerous to who???
To government hierarchs, whether elected or un-elected?
Sure!!
It is meant to be that!!
The usefulness of government hierarchs to the Civil Society they tax is quite limited. The encrustation on Civil Society of so many politically allocated programs since Apollo 11 sparked calls of, “If we can go to the Moon, …then why can’t we, …..”, is a dragging weight on a highly productive Civil Society otherwise following paths un-productive of votes for politicians.
There will *never* be a Moon Settlement, or a Mars Settlement, or anywhere else, that is funded by Congress, or any other political body outside the reviving “Central Kingdom” in Beijing. Probably not even there. Trying to preserve the encrustation of politically allocated resources by saying we must be cautious in our praise of a rapidly innovating portion of Civil Society adds little to the conversation that does not boost politicians.
Not helpful in terms of how to understand the problem and fix it.
IMHO, the problem is inherent in the behaviors found in all political hierarchies. The top levels of those hierarchies have agency costs that cannot be mitigated. The key is, as much as possible, to do without the hierarchy. If you cannot do without it, then minimize the number of layers in the hierarchy that insulates those at the top from accountability for the disruption of the hierarchy’s purpose, that their agency costs cause.
Yeah, this tweet reeked of run-of-the-mill SpX fanboyism. Just last year we were hearing ‘SPACEX IS LEAVING NASA IN THE DUST. NASA IS STILL DESIGNING SLS AND SPACEX IS BUILDING COMPOSITE BARRELS’
Then November and SpX decides to do it in steel, no one said a word when it became obvious that a changing top-level architecture meant that anything below that architecture didn’t exist.
Fast-forward to now that SLS is pretty much in final assembly and test and SpaceX has a big steel grasshooper. ‘WE HAVE A FLYING VEHICLE AND SLS IS STILL IN PIECES’
But this is beyond all of that, like you said. NASA is a space agency, not a company to root for.
OK. Where do I start “natedogg” : NASA has been building SLS for how many years – with the first flight delayed again and again. SpaceX built Falcon 9 and Falcon Heavy in less time. And their rocket is reusable and vastly cheaper to fly even if you include every penny that went into building it. The Star Hopper did not even exist a few months ago.
No, changing a high level decision doesn’t mean they hadn’t done any more detailed work. I suppose it can, and we’ve seen that more than once for things like human spaceflight to the Moon or Mars. But it can also mean someone _did_ do the more detailed work. Enough of the more detailed work to realized the high level decisions were based on bad assumptions (e.g. how hard or easy it would be to build large composite tanks.)
When you find out your original plan is not going to work nearly as well as you expected, changing your mind and going with plan B is a sensible thing to do. Proceed as originally planned, no matter what, or cancel isn’t a healthy mindset.
LOL. What you’re missing is the fact that SLS will fly at most twice per year and it’s got a program cost of $2+ billion a year. One flight per year is actually a more realistic goal considering its expendable nature and the issues that keep cropping up. So say $2 billion a year for one launch. Is that really worth it?
Starship/Super Booster, on the other hand, is being designed as a fully reusable TSTO. Even if it fails to meet Musk’s expectations (turnaround measured in days with little to no maintenance), it will still be far more economically sustainable due to reuse.
Do you have some examples for “NASA has hoards nipping at their heels, each with a smug sense that as a citizen they can do a better job.” and “while NASA failure is a waste of taxpayer money, or more pork, or whatever”? I’m not seeing it. If NASA is doing innovative work and fails, I think most people would understand and support them, the problem is they’re not doing enough innovation.
I actually can’t think of any recent examples from NASA, but there’re plenty of examples from DARPA where they tried to do something crazy and it didn’t pan out, ALASA is an example, I don’t think there’s much negative reaction.
“the problem is they’re not doing enough innovation” – I agree with this statement. Year after year we’re seeing innovative R&D being cut at NASA to feed programs like James Webb Space Telescope, SLS, and Orion. And don’t get me started at the pittance of money allocated on the aero side of NASA. That’s the bit that’s supposed to keep our domestic aircraft industry on the “bleeding edge”.
What SpaceX is doing in Texas is validating the Raptor engine by putting it thru some real world practical testing. First one , then three engines , firing longer and going higher. You cannot really say that is a bona fide rocket booster they are cobbling up and making it dance. It’s a boilerplate proof of concept engine test stand that greatly resembles a Buck Rogers rocket , a minimalist derivative of the planned BFS spaceliner with respect to size shape and weight and not much else . I note there is no flame trench visible. This Hopper is gonna roar to life and flap its fledgling wings off flat ground from the looks of it.
Having said that , I see no reason why SpaceX cannot stay with this drydock shipyard model and build a true suborbital booster in situ out in the open , and fly it. Seems to me they are putting the first few BFS together from modular components with greatest of emphasis on simplicity and robustness. True grit. Reminds me of how they must’ve built clipper ships and schooners in the yards of western Europe during the colonial era… design as you build. Then go to sea and do vast commerce abroad. In this case, up.
Boeing , Lockheed , and Orbital Grumman Aerojet Acme Whatever are being schooled in 21st century space transportation and off planet colonial expansion. I hope history has some bearing on all this.
Yes, Elon Musk is going to fly it through tests similar to those done by the DC-X. Meanwhile he has already started building the full size Starship.
Yeah, it’s more like a free flying (tethered?) test stand, if you will… 😉 I imagine the purpose is to test the raptor engines and control systems in a dynamic environment, rather than to test full flight-like systems integration. The shiny wrapper and erstwhile nose cone… marketing.
Maybe marketing, depending on your POV, but Elon has said and otherwise hinted that they needed the mass, and the appropriate measures of mass (center of gravity, etc) to replicate the final spaceship, from which one could figure that even a dummy would have the appearance of the final spaceship. Or not?
I haven’t seen Elon’s comments regarding that. Replicating things like CG and moment of inertia doesn’t require similar shapes, although it may help. Perhaps if it were for testing aerodynamic properties… but if that were the aim, I’d expect to see a scale model prototype, which this wasn’t. In any case, it only took them a few weeks to build the nose cone, and must have been relatively inexpensive. Given that they choose not to replace it, I’d imagine whatever test value it may have had must have been quite limited. If I were a betting man, I’d bet it was just cosmetic.
At this point, I will take the off-ramp marked ‘Liberal Arts Majors, Exit Here.’
If the vehicle is going to be climbing at 100m/s or more aero drag would be significant and the fairing would help. It’s definitely not going supersonic with a flat nose. If it is just for low speed tests it may not make a difference. Ballast can be added if needed, and a lot of data on throttle and thrust vector control system performance and reliability can be gathered without duplicating the aerodynamic forces, some data on fuel sloshing can be tested as well.
Keeping the engines and control systems exposed in the beachside air for weeks is going to be a pretty rigorous test in itself, but at some point they will need an assembly building if just to keep out sand, and some assembly fixtures to get the dimensional precision needed for an actual launch vehicle.
It is hard to see how they can assemble quarter inch plate to tolerances adequate for spacecraft by welding without assembly fixtures. The plates seem to be held by guy ropes. Common practice for ground-based welded storage tanks but they have higher design margins and safety factors than spacecraft. Consider the Atlas I – the last big rocket made of stainless steel. It worked well but the steel is quite thin and the welds must be precise.
Meanwhile SpaceFlightInsider.com is reporting that Boeing’s Starliner schedule may slip up to 6 months.
From the ‘Better Late Than Never Department’, I noticed this piece today:
https://www.space.com/nasa-…
In which is described NASA’s (very) small additions to spacecraft from India and Israel in the course of a few weeks.
Loosely apropos “better late than never”, I saw Apollo 11 last night. Glorious. If you haven’t seen it already, do so.
Weeks impresses me. My first reaction was that those are visible light (laser) reflectometers, and that’s almost the simplest instrument I can imagine. It’s completely passive and you can make one from three mirrors bolted together at right angles. Then “weeks” sunk in. Even for something simple like where the bolts and bolt holes go, to mount something on a spacecraft, NASA would normally to want a mechanical interface control document. That isn’t something that typically gets agreed on, written and signed in a few weeks.
“Weeks” is probably a function of the Israeli and Indian ability to quickly integrate.
Or possibly because the Israeli and Indian project managers just decided it was their mission. I mentioned interface control documents as one thing which can get in the way of fast work. That’s because they are usually a negotiation between the instrument team and the spacecraft team. The host (spacecraft team and manager) could simply say, “This is what the interface will be like, if you want to fly on our vehicle, follow those rules. No discussion, take it or leave it.” If that happened, I could see it speeding things up by a fair amount (especially for a simple instrument like this.)