Starship SN9 damaged in an accident in the high-bay | Hacker News Reader

Starship SN9 damaged in an accident in the high-bay

twitter.com

43 points_Microft56 comments

The sad thing is that SN9 was meant to be moved to the launch pad on monday next week already. This is pretty certainly not going to happen as one of the rear flaps is damaged.

Maybe assessing and fixing the damage would not even be worth it since SN10 and successors are being built right now? (Parts belonging to prototypes up to SN16 have been spotted by a local SpaceX fan already.)

> Maybe assessing and fixing the damage would not even be worth it since SN10 and successors are being built right now?

Treat your starships like cattle, not pets!

Fly it anyway and do a DiRT (pun intended).

Likely the most valuable parts are any engines that were mounted. (Have they mounted the engines?) My guess is that if they have a current surplus of engines, they'll try a fix and attempt to get data out of sn9, even if success chance is low. If they have more pressure vessels than engines for those pressure vessels, then I doubt sn9 will fly.

Either way, one of the larger goals of these early prototypes is to iterate on how to make the starship quickly and inexpensively, so I doubt it's a total waste even if it doesn't fly.

Some see this as a complete failure, but it does also provide input into the practices that led up to and caused the accident. Rocketry is not limited to the activity between the launch and landing pads.

This is precisely the reason there are so many tests - there WILL be failures along the way. There are at least 7 more such tests in the queue already.

I know this is probably old hat by now but that thing looks massive with that person at the bottom for scale, especially considering it's only the upper stage.

Stainless steel. It'll buff out.

Is it really clad in stainless steel? Seems like there would be lighter materials that would serve that role.

There are lighter materials but apparently they're more expensive and harder to work with, so steel ended up winning the cost/benefit analysis.

There's been ample discussion about that when they originally started talking about it, e.g. https://www.space.com/43101-elon-musk-explains-stainless-ste...

I think the real reason is Elon Musk wants Buck Rogers style spaceships and will do whatever necessary to get there.

Probably not, SpaceX invested a lot of time and effort into composite tanks before giving that up.

Having product that appeals to his not-so-hidden inner thirteen year old boy is definitely a motivation for Elon Musk. Exhibit A: cybertruck.

Actually getting to Mars is a higher priority, though.

Exhibit A', flame thrower.

It's not clad in stainless steel -- it _is_ stainless steel. The reason being it's cheap, and that especially the alloy they use is very robust under cryogenic temperatures as well as high temperatures.

Yep, you can watch videos of the assembly on YouTube by Boca Chica Girl and see all the components. Stainless through and through.

That's the problem with 'seems like'

At room temperature, yes. At the range of temperatures this needs to operate at, stainless is actually lighter for the required strength.

Keep in mind that this is a tank for cryogenic propellants (almost-solid liquid oxygen and methane), that needs to survive atmospheric re-entry from interplanetary speeds, so it needs to be strong in temperatures around -200°C to +400°C.

I realize there's technical reasons/tradeoff's for using stainless, but goddamn it looks cool.

It will be even more magnificent if they can make the skin taut. Isn't that the intent in future models? Though the sheet metal waviness is nice in a retro way too.

It needs to act as both a tank wall for cryogenic liquids that won't crack and as a backing for a heat shield that won't melt. Tough trade-off.

Melting point of aluminum is too low. Burning point of epoxy/carbon fiber is also probably quite low, but it's also famously hard to work with and get reliably good results that you can actually inspect. Titanium might work, but is also had to work with and tends to crack.

How do accidents like this even happen at an established company? That sort of stuff indicates a worrying proclivity to cut corners.

All of this work is designed to be highly iterative, (relatively) low cost prototyping. Cutting corners is exactly what they are trying to do. If you hadn't seen before, until SN8 they were flying this thing without a nose cone. Cutting corners here is OK. The point is to find the mistakes. Better to find these issues now than to be something like https://en.wikipedia.org/wiki/NOAA-19#Damage_during_manufact...

Cutting corners here is OK

No, it's not. Incidents like this tell a lot about internal culture. Things in highbays aren't supposed to fall over just like that, that sort of accident could have cost a life and certainly has lost the company money and set it back a few months. Safety is very clearly an afterthought at the Starship site. You can't have that in rocketry. They want to go to Mars and can't even get out of the assembly building in one piece!

Starship is basic engineering in metal. The thing sat there and just fell over. There wasn't even any kind of load on it. That takes some effort!

I tend to agree with you, and that was definitely my first thought. However we don’t know what was going on when this happened. If it was just sitting on the stand and fell over that’s a problem, but they were clearly getting ready to move it and may have been transitioning to a set up where they could lift it. It’s possible (albeit unlikely) that this was a risk that they had calculated and put some countermeasures into place for.

They just dropped one of these out of the sky from 8 miles up, but since the risk was managed in a very visible way it generally wasn’t considered a safety issue.

This is a prototype during R&D. It's like when you press F5 and you get a compiler error because you accidentally used `.` when you should've used `->` to access a member function.

Does this tell a lot about your abilities as a programmer? Do you hide this incident from your boss so that he doesn't realize you're a fraud? Should you tell your customers about this? It'll be much worse if they hear about it from someone else.

The difference between `.` and `->` is basic engineering! You're trying to ship a product, but you can't even get out of the compiler in one piece!

Some disciplines just don't lend themselves to the "rapid prototyping" mindset. Imagine a process engineer scaling up a chemical reaction in the pilot plant and neglecting the cooling capacity of the reaction vessel. A thermal runaway is encountered, and the vessel contents, 100s of liters of flammable organic solvent are distributed throughout the building.

The fire chief will want to have a word with the rapid prototyper, sizing the thermal budget has been practice for 100 years.

Software engineers have an odd mindset towards bugs. It's already an outrage that when you activate a new cellphone the first thing the device does is download gigabytes of updates. It's a disgrace.

At the same time, out of the rest of the new super-heavy launchers under development, New Glenn, Vulcan, SLS, Ariane 6, and Long March 9, none have flown. Yenisei is expected in 2028.

Meanwhile, SpaceX continues welding sections of their new rocket in tents in the Texas brush land, and days ago performed one of the most impressive tests in recent memory.

But thats not prototyping... thats testing in production. Two different things.

Prototyping would be an engineer in a lab with a scaled down version of whatever process he is working with. He tries 10 options, 9 of which blow up. But since its scaled down the blow up is just a little fizzle, so who cares. The 10th option that worked can be developed more and eventually brought to the full production plant

> set it back a few months.

I'd bet $1000 that by end of next week SN10 will be on the pad, if not a repaired version of SN9.

> Safety is very clearly an afterthought at the Starship site.

As far as I know no one has died at a SpaceX site. I think clearly we can say that SpaceX is putting sufficient amount of energy into worker safety. I agree with you this issue posed a risk to safety, and I'm sure SpaceX will learn from it and applies the lessons from it in the future.

> You can't have that in rocketry. They want to go to Mars and can't even get out of the assembly building in one piece!

Literally one day before they just took almost the same rocket up to 12.5k and almost landed it. I'm trying very hard to assume good intentions on your part, but statements like that make it hard.

The alternative you seem to be proposing is the Boeing model where everything is extremely slow, methodical and waterfall style. How did that end up? A disaterous demo flight with 2 near misses: https://en.wikipedia.org/wiki/Boeing_Orbital_Flight_Test#Ano....

We all agree rocket science is hard. We all agree that going to mars has lots of critical safety requirements. But SpaceX isn't going to mars yet, they are just doing rapid prototyping. The requirements here are WAY less.

Unfortunately the number of deaths at SpaceX sites is non-zero.

I do not know details of the accident today, but there has been a death in the past.

(2014) https://www.spaceflightinsider.com/organizations/space-explo...

The next road closure is December 28. I don't think it set them back too far, but a week was overly optimistic.

You've been downvoted but clearly heavy objects falling over in workshops where workers might be squished is not meant to be part of their corner-cutting rapid prototyping approach to rocket design. That's an industrial accident, not prototyping. Rockets blowing up on launch pads is fine, that's to be expected. But that's not what this is.

I just wish some of the downvoters would give their reasons. To me, the Starship project is unusually prone to safety incidents, which tells a lot about organizational culture.

This is the first incident I've heard of that seems concerning, but admittedly I've not followed the matter closely. I've heard they had some incidents on the pad, or the one they crashed earlier this week, but neither of those concern me since failure in those scenarios is obviously something they plan for (nobody would be nearby.)

I found that tank testing incident where a fuel tank exploded during static pressure testing below rated pressure extremely worrying. Pressure vessels are well understood, they aren't supposed to fail like that. Don't they examine their welds at Space X?

Such a failure seems reasonable to me, such a pressure vessel obviously needs to be as light as possible to maximize payload. But making it lighter will make it weaker. And even today there are limits to computer analysis, hence why they would do pressure tests at all; if the models were foolproof then why would anybody test pressure vessels? Pressure vessels are tested because their danger is well understood. Testing would be a waste of time if outcomes were certain, so failure during tests is not concerning.

(This rocket falling over was presumably not part of a test to see whether it will fall over.)

I mean, that's why they test? Can one presume that there was an improvement based on that outcome?

I didn't downvote you but you make a lot of patently false claims, which at best show you're ignorant of what SpaceX is doing. As others have said:

- You say "some domains don't lend themselves to rapid prototyping" - yet SpaceX's results speak for itself

- You say "unusually prone to safety incidents" - can you list 5 of them so that we know what you're talking about? "testing incident where a fuel tank exploded" - that's literally their designed operating mode, nobody was put in harm's way, you're basically disagreeing with how they operate (should have the humbleness to consider that their impressive results so far might indicate they're right and you're wrong)

- You say "they can't even take a rocket out safely, how are they hoping to achieve anything" just after even their competitors praised them for the impressive results.

etc. Is it really surprising that you get downvoted?

It set it back 11 days apparently... not a few months. They’re moving it to the launch site now: https://twitter.com/BocaChicaGal/status/1341378205962395649?...

"Repairs to the satellite cost $135 million. Lockheed Martin agreed to forfeit all profit from the project to help pay for repair costs" -- yikes! Imagine being the poor sap responsible for removing those bolts. O_O

This is after Lockheed Martin's error lead to the loss of the Mars Climate Orbiter. They shouldn't be allowed near any spacecraft ever again.

https://en.wikipedia.org/wiki/Mars_Climate_Orbiter#Cause_of_...

We could always do it the NASA way and.... just talk about it for 40 years and then spend double on something that doesn't work...?

:-)

Falcon 9 launches customer satellites. It is very reliable.

Starship is an experiment. It explodes a lot.

SpaceX is eating the entire industry because they understand the difference.

The fanboism about SpaceX is beyond belief! Like they are the only private enterprise that works in this area! For example, there's no excitement here on HN about competitors such as Relativity Space [0], which has some really unique solutions! Or the many others in the field. Every failure of SpaceX is considered here a huge success!

[0]: https://www.relativityspace.com/

There are a lot of companies like that popping up now in the wake of SpaceX's success. Enough to fatigue most peoples' capacity for excitement I think, particularly in the cases where the company has interesting ideas but has not yet put anything into orbit. SpaceX isn't just talking about paper rockets, they're also putting stuff into space regularly for the past several years. It's hard to compete with that for attention if all you have is neat ideas and a CEO who isn't a media hungry controversial celebrity billionaire. (Blue Origin almost has that, but I think they are too busy being secretive to nurture an army of fanboys.)

I'm totally ready to be a BlueOrigin "fanboy" but their bigger promises haven't materialized yet. I think Rocket lab is cool and follow their progress, but their ambitions are comparatively small.

A lot of rocketry advancement would not be happening if SpaceX didn't enter as a competitor. For that reason alone SpaceX should be lauded. It helps that SpaceX is pursuing the dream of a space faring civilization that was promised decades ago.

Relativity Space looks cool. But until they launch they will sit in the stack of neat tech to maybe be excited about.

I am excited by humans doing things to advance the species. I don't care much who is doing it. But, I have been burned by too much vaporware to get excited by marketing pieces.

If you haven't watched Smarter Every Day's ULA factory tour I highly recommend it. There is even a nice bit about how their materials have improved as computers were better able to model stresses. https://youtu.be/o0fG_lnVhHw

Maybe, I can't tell if that's good or bad, but I also think all mindshare goes to SpaceX and it's not a good thing. And, also, can you compare the character of Tory Bruno to Elon Musk's? It's like comparing a day to a night!

There are various reasons for why SpaceX is so popular, but I think most important is because of Elon starting from scratch against incredible odds of failing, combined with massive resistance from established companies and initially even the government (to the point of Elon crying during an interview), combined with Elon's novel and bold designs/goals and SpaceX's engineering feats. Add to that also their willingness to be open and share success and failure with the world, and I think it becomes easy to understand why they have so many passionate fans.

Sorry, can't sympathize. When I compare Elon Musk to Tory Bruno, it's a vast difference. Tory truly is a rocket man!

I am a fan boi of cheap reliable access to space. It just happens that SpaceX is the only game in town.

Relativity Space only has paper rockets at the moment, and their best design is 4 times more expensive per kg than the falcon 9.

The only organisation right now that is working on a rocket that will compete with the falcon 9 is Blue. No one is working on something that will compete with starship.

Remember when people said this about Tesla? "They have tents outside to build in. A disturbing corner cut". Bam! Better electric car than anyone else.

Reply on news.ycombinator.com