Starship is Still Not Understood (2021)
caseyhandmer.wordpress.com
caseyhandmer.wordpress.com
Again, this is from 2021, one of those set of launches was to be last year. And it was to be 1,000 starships bound for Mars to meet that goal.
Starship is impressive but we need to be realistic about these numbers. We haven’t even technically had a launch with any payload. Additionally they claim that within 2 years of the Artemis mission we could get 1,000 people on the moon (though say this isn’t going to happen at least).
Starship will make Starlink V2 feasible and dozens if not hundreds of missions that would otherwise never be possible. But getting 1 million people to Mars (or even 1,000 people to the Moon) is just fantasy. I will gladly eat crow but none of these capabilities that exist on paper have actually been demonstrated that are necessary to meet these goals. SapceX is the best there is and likely will get to Mars. Just not how Musk describes it.
Arguably he tweets about it. For the most part the 'doing' of the cool stuff is happening due to the Engineering and Manufacturing teams who are putting in sweat equity on that vision.
Judging by the erratic and deteriorating public displays in the past few years I am not even sure he does anything on the strategic level anymore.
"Why don't we try to use the tower to catch it?" he [ELON] asked. He was referring to the tower that holds the rocket on the launchpad. Musk had already come up with the idea of using that tower to stack the rocket; it had a set of arms that could pick up the first-stage booster, place it on the launch mount, then pick up the second-stage spacecraft, and place it atop the booster. Now he was suggesting that these arms could also be used to catch the booster when it returned to Earth.
It was a wild idea, and there was a lot of consternation in the room. "If the booster comes back down to the tower and crashes into it, you can't launch the next rocket for a long time," Bill Riley says. "But we agreed to study different ways to do it."
A few weeks later, just after Christmas 2020, the team gathered to brainstorm. Most engineers argued against trying to use the tower to catch the booster. The stacking arms were already dangerously complex. After more than an hour of argument, a consensus was forming to stick with the old idea of putting landing legs on the booster. But Stephen Harlow, the vehicle engineering director, kept arguing for the more audacious approach. "We have this tower, so why not try to use it?" After another hour of debate, Musk stepped in. "Harlow, you're on board with this plan," he said. "So why don't you be in charge of it?"
https://x.com/WalterIsaacson/status/1844870018351169942/phot...
Every innovative mission needs someone pushing people to do the impossible. That being said, every company trying to accomplish the impossible needs an adult in the room to ground that visionary.
In the case of Space-X, there's an adult in the room. Elon's direct input at Space-X almost set the entire Starship project back a year by insisting on trying a concrete launch pad that all of his engineers told him would fail (and it did). After that failure he appeared to get so distracted by Twitter that Shotwell could point the engineers in a more effective direction.
In the case of Tesla, there unfortunately is not. Elon's direct input resulted in canning a low-cost model in favor of a "robo-taxi" that is 100% dependent on full self driving using nothing but cameras. That's pretty obviously not going to be viable with any existing technology. An adult in the room would've told him to either stick with the low-cost model to compete with incoming Chinese imports, or use Lidar so that there's a viable self-driving vehicle that can be sold sometime in the next decade...
It may be that this layer of management is getting weaker.
Musk's biographer tweeted the pages from his book <https://x.com/WalterIsaacson/status/1844870018351169942> discussing how in late 2020 Musk suggested, then insisted against considerable opposition from his engineers, that Superheavy be caught with chopsticks instead of landing on legs like Falcon 9.
(If this sounds familiar, also according to the book, Musk is the person who suggested and, against considerable opposition from his engineers, insisted on Starship switching to stainless steel instead of carbon fiber.
Hint: Musk was right and his engineers were wrong. Both times.)
The lunatic alone will hype and then fail. The sane voice alone will plod along and take aeons to make any progress, or stagnate entirely.
SpaceX is just spectacular. Something is different there.
Shotwell is probably a big reason SpaceX does better than Tesla or the others. Elon needs to be paired with a cooler head that knows how to manage well. I get the sense the others don’t have anyone as strong in that role.
It’s easy to shout from the sidelines that you’d do better given the opportunity, but reality suggests it’s more difficult. It’s the HN equivalent of sports fans hurling abuse at their football team’s players.
Bill Gates eradicated polio. Sam Walton re-invented shopping at scale. Jeff Bezos re-invented shipping logistics.
I know you reeeeealy want to defend Musk, but think about your argument before posting it next time.
Counterpoint, new technology needs a bleeding edge product necessitating and thus incentivizing its birth and existence.
When you need to create a chicken from nothing so you can get eggs which turn into chickens for more eggs, it's oftentimes expensive as fuck and more likely than not going to fail. But someone needs to do it if you want those eggs and the chickens.
He also came up with building it out of stainless steel, which worked, and catching it at the launch tower. But I'm sure you know better.
So you didn't follow it at all? Because multiple government agencies were considering blocking them full-stop for an extended period of time after that disaster. So "maybe 2 months" was their absolute best-case scenario.
https://www.tpr.org/environment/2023-04-27/photos-spacex-is-...
>He also came up with building it out of stainless steel
And you know he was the one who personally came up with that idea... how? You work at Space-X? Or you've just read a couple press releases where he took credit for it?
Elon has a long history of taking credit for other people's ideas. And I'm willing to bet whatever you'd like to risk that one of the engineers at Space-X suggested using stainless steel, not Elon.
Tell me you’re an Elon apologist without telling me you’re an Elon apologist.
The current regime of tariffs and restrictions means they won't have to.
What Musk understands is that technology is a moving target. You shoot for where it will be, not where it is now. Sometimes he's overoptimistic, however a robotaxi seems well timed.
Have you seen the GPT-4o videos? I don't know how anyone can watch those and believe that vision e2e self-driving vehicles are somehow a long way off.
It takes years to manufacture a new vehicle type. When the Cybertruck was announced the SOTA in language models was GPT 2. Now we have multimodal conversational models with vision.
By the time the robotaxi rolls off the production line, the software will be ready. And if the software is ready, then the model of owning cars will be disrupted for many people living in dense urban environments.
The real risk is in not preparing for that eventuality, which is the same type of unpreparedness risk that the legacy launch industry is facing with Starship.
Because chatting and driving a car are two entirely different problems?
> ...AI is evolving quickly in many domains, so it's not a huge stretch to think we might make another order of magnitude improvement in self-driving cars. At least enough improvements to fully solve the robotaxi problem.
But that's sloppy and fallacious reasoning, and there are literally thousands of examples of it failing in many different areas (including AI!) in recent and no-so-recent history.
One example is SHRDLU, which led to similar excessive optimism back in the late 70s: https://en.wikipedia.org/wiki/SHRDLU#Consequences.
It's like someone with the fuzzy category "technology" interpreting the amazing advances in computer graphics over the 90s and 2000s to conclude we will be colonizing Mars soon.
We need to be realistic about timelines. Specifically, the timelines promulgated by Musk versus SpaceX. (Not an easy task, I know, in part because the latter is more sparing with such predictions.) But we should also be clear eyed about emerging capabilities.
Barring a major engineering challenge in in-orbit propellant transfer, Starship's fundamental design looks solid. V1 probably won't get a million people to Mars. But V1 Falcon couldn't be re-used twenty-odd times. The author is correct: national space agencies are copy pasting legacy missions onto paradigm-changing platforms.
The closest exception is NASA: Artemis is overly complicated for a single Moon landing, but its use of Starship on the Moon and for in-orbit re-fuelling is a bet-the-house play that could open a real space economy in propellant alone. (First, storage and shuttling. Later, manufacture.) To the degree Artemis might get canned, it's because Lockheed can't manufacture an Apollo-era heat shield [1], Bechtel can't build a non-catching tower [2] and NASA refuses to be mean to them.
[1] https://spacenews.com/nasa-inspector-general-report-highligh...
[2] https://oig.nasa.gov/wp-content/uploads/2024/08/ig-24-016.pd...
We already have that challenge. Last I checked, SpaceX was claiming it would take five Starship launches to store enough fuel in the depot to perform one lunar mission (which is innately strange considering Apollo did it in one launch on a less powerful rocket, but that's not even the point here). But according to NASA, SpaceX has failed to adequately account for cryogenic boil-off in their plans, meaning that they expect it will take about twenty Starship launches per lunar mission. That's not exactly inspiring, especially given what it implies about the number of launches it would take to get to Mars and back. To solve the cryogenic boil-off issue we might need to switch to different fuels entirely, which would necessitate new engines.
Apollo was disposable. Starship is not. Apollo could thus max out payload at launch, which Starship cannot do design or fuel budget wise.
The Apollo paradigm was for literal moon shots. Starship and Artemis are more about building infrastructure.
> according to NASA, SpaceX has failed to adequately account for cryogenic boil-off in their plans, meaning that they expect it will take about twenty Starship launches per lunar mission
That is not NASA's position. Folks at NASA and SpaceX disagree around the boil-off rates. Everyone thought about it. We just don't know what it is. We should have a preliminary figure next year, and a final-ish one within two or three years (after the first vehicle's boil-off characteristics are optimized).
But yes. There are major engineering challenges in-orbit propellant transfer could raise that could require going back to the drawing board. Hence why I flagged it. Given that's something humanity has never really done, it seems like a fair risk to take given the upsides.
(Note, too, that in present designs Starship must function as a launch vehicle, spacecraft and fuel tank. That's not required long term. Well-insulated orbiting fuel depots start making sense rather quickly if you have <$1000/kg launch costs.)
https://spacenews.com/starship-lunar-lander-missions-to-requ...
Of course this depends on FAA permission. Whether they get FAA permission may well depend on who wins the election. Musk has gone out of his way to place a big bet on Trump, and it is clear that Trump will happily get rid of red tape for Musk in return. My read is that part of the reason why Musk placed that bet on Trump is because the Biden/Harris administration has already decided to place lots of red tape in Musk's way. At first as a reaction to Musk's hostility towards unions and COVID restrictions on manufacturing, and then as a reaction against what Musk was doing with Twitter.
You're not launching exquisite rovers or satellites -- you're launching a bunch of fuel.
So if a reused Starship blows up on accent*, that's not great, but it's not terrible either. Load the next one and try again.
* Obviously, you want to be much more careful on rendezvous
You're correct. But Starship, being a multi-use spacecraft, can't capitalise on this differential.
Fuel-only launch enables high-g launch options, e.g. rail guns. It could also enable low-reliability low-cost options, but the nature of space launch hasn't opened that niche up, and--I suspect--the nature of environmental regulations ensure it never will.
And likewise, probability of relaunch success scales with the money and time spent on those other two.
In fact, I'd strongly argue that fuel-to-orbit is Starship's version of Falcon's "Starlink demand seeding."
It's win/win if SpaceX has a high-volume, low-payload-cost, revenue-generating demand for Starship, as it gives them a failure tolerant mission to push relaunches to failure and then use that data to improve future reuse reliability.
Of all the things SpaceX has done, getting themselves into positive-feedback flywheels has been the smartest.
According to one person at NASA. There's disagreement on the point.
https://arstechnica.com/space/2023/11/what-nasa-wants-to-see...
"[Lisa Watson-Morgan, who manages NASA's Human Landing System program] suggested the range in the number of Starship tanker flights for a single Artemis mission could be in the "high single digits to the low double digits." Elon Musk, SpaceX's founder and CEO, has suggested the company is looking at expanding Starship's capacity with larger propellant tanks. This could help reduce the number of tanker flights."
There's a lot of unknowns. Falcon 9 versions saw significant performance boosts over the generations; we may see the same with Starship.
Conclusions:
* Standard Starship is probably good enough for HLS
* Simple coatings and/or a specific orientation makes things a lot better
Literally every expert said that about landing the first stage of Falcon 9.
Then they said it about reusing it more than 10 times.
Then they said it about having good internet from a mega constellation of LEO stas.
Then they said it about using stainless steel for starship.
Then they said it about having a full-flow staged combustion cycle engine
Then they said it about catching the booster with chopsticks.
Now they're saying it about...
Yes, Elon is wildly optimistic about timelines. I have to laugh at his quote "At SpaceX we make the impossible merely late".
You can't deny SpaceX are absolutely doing what everyone said was impossible, and they're continuing to push forward. You'd have to be nuts to bet against them.
UPDATE: It was only a couple of months ago the CEO of United Launch Alliance called out SpaceX for showing what he claimed was a "partially assembled engine" [1].
It was a fully-functional engine that test fired [2]
Even the CEO of their biggest competitor, and ostensibly the second best rocket company on earth can not fathom what SpaceX are doing.
[1] https://x.com/torybruno/status/1819819208827404616
[2] https://x.com/Gwynne_Shotwell/status/1821674726885924923?t=v...
That would have involved launching 1,000 Starships last year. They have not done *1* launch with 100 tons of payload, the capacity of this rocket on paper.
They will absolutely with this design get to Mars. But the realistic timeline is within the next 20 years, and only a few hundred tons. It will be a monumental achievement that will be immortalized for the rest of humanity. But that scenario is 1/1000th of what Musk describes.
Starship is an incredible program but it is still rocket science not rocket fantasy. Please don’t think being realistic about getting 1,000 Starships to Mars in the next two years means I think Starship is a failure.
Elon Musk launched the first mission to Mars in 2046.
Nobody cares about delays.
But I mean, most of us here are probably the same. Most CEOs I’ve worked with are quite poor at time estimates too. Come to think of it, I’ve never met a single human who is good at it.
The problem is often unknown unknowns that you had no way to predict. I once had a project delayed by months due to a rare bug in the .NET CLR that just took a really long time to track down. I don’t know much about rocket science, but operating at the edge of known science and technology presumably has some difficult periods...
Hofstadter's Law: It always takes longer than you expect, even when you take into account Hofstadter's Law.
Yes. Everybody knows.
With SpaceX we’re late getting there. With Everyone else the timeline is never.
I’ll take late over never 10 times out of 10.
I think an important part of this quote is “At SpaceX”. For my part, I generally assume that “SpaceX Elon” and “other venture Elon” are different people. He has certainly burned a lot of good will with his FSD predictions, and to a smaller degree with things like hyperloop, boring company, etc.
SpaceX in contrast has always greatly exceeded expectations.
Any answers shouldn't consist of "well it's a fraction of what the aviation industry emits" because _the entire point_ of Starship is to scale-up to aviation-like operations.
Your asteroid mining for profit is arguably meaningless if we have no ozone layer.
Now high-altidude effects would increase the greenhouse effect of these emissions by some factor I don't know. At the same time a large amount of fuel is actually burned by the booster at below airliner altitudes. Another mitigating effect is that some of this methane could be produced renewably.
I haven't heard about Ozone depletion from spaceflight except for the case of reentering Aluminum burning up. Well this wouldn't be an issue with Starship flights.
So you know overall, let's just give it a lot of conservative margin and imagine we are looking at about 10% of the US airline or 0.2% of total US greenhouse impact. That is on the scale of a whole new industry, yes, but if you can imagine a world which has attained a sustainable rquilibirum surely you can also imagine one which has a bit of margin left? And this is only for the US!
What do you gain? Three of the points I mentioned have potential direct benefit for people.on earth: asteroid mining, L1 solar shield and brilliant pebbles.
Asteroid mining "for profit" also means that the economy profits, unless you believe that capitalism doesn't work. I read that less than 200t platinum are produced annually. Say your asteroid mining crashes that economy. I found some figure of 20kt of CO2 for each ton, so 4mt in total. That is more than your 1000 launches emit! Now my calculations are of course wrong but it seems plausible that the orders of magnitude could match. So you should definetly consider that your asteroid mining operations might stop other mining operations from happening which reduces your greenhouse impact.
Then you have the solar shield. It is certainly possible, although of course difficult and has some risk. But it's good to have a backup plan, no?
Mentioning theoretical solutions doesn't help if they will never be used practically.
You want more specifics, I take it?
* It's been well known for decades that high-altitude supersonic flight damages the ozone layer significantly faster than conventional aircraft. New Scientist, 1997 [1]
* More recently, concerning reentries of spacecraft: Atmospheric impacts of the space industry require oversight, Nature, 2022 [2]
Our upper atmosphere is fragile and contains the bulk of the ozone layer that protects us all from skin cancer and being bathed in brutal ultraviolet radiation. You willing to risk it because "space is cool"?
[1]: https://www.newscientist.com/article/mg15320692-500-science-...
If you are an engineer or even have half a brain nothing prevents you from doing a more in depth analysis, make a blog and post it here.
Right now not only are you a naysayer but you are even putting on hacker news commenter the burden of defending your point.
For the first study, you have conveniently focused on "supersonic flight" as the issue which is not the fundamental issue cited in the article (link isn't working for me, btw, I had to google it find it myself). The fundamental issue is in aircraft fuel and the sulphur based impurities it carries. Starship burns Methane and LOX. In general, these fuels (fuel and oxidizer) are typically kept as pure as possible and are unlikely to contain much more than trace sulphur impurities as there removal is a standard and well followed practice.
So for full clarity, your argument as presented here is completely without merit and it looks like your first study was intentionally presented in a way to obfuscate its extremely tenuous link to the issue at hand.
As an aside, you're also quite focused on your mood affiliation with the topic, doing things like asserting the motivation of those you reply to without any evidence (eg 'You willing to risk it because "space is cool"?'). This won't help you think clearly about a topic.
the JPL produces things at $1,000,000/kg, and that inside the engineering organization it'll be hard to reconfigure everyone's mindsets and practices to create stuff that doesn't spend so much money / is less concerned with weight.
From an engineering perspective it seems like simply putting less into a finished product would be easy, but his argument is basically that the design parameters for every single thing produced are so fundamentally oriented towards weight (and size maybe?) that it's very hard to turn around. (This is not exactly the government spending / misaligned incentive argument that is made for many government projects- my read was that this is a subtly different problem.)
It sort of is. The folks selling specialised design services for optimising the third sigma of weight savings absolutely don't want NASA to use off-the-shelf components.
Does he mean grid fins? Because "flaps" are something else entirely.
Grid fins have been around since the 1950s.
Anyone know what it is now? What it will/would be with Starship?