Hacker News Reader

Comment by cryptonector | Hacker News Reader

> And right now, the soonest likely to launch is being led out of a sprawling former Fry’s Electronics retail store in a shopping center complex in Texas.

Oh man, I miss Fry's.

Anyways, ISTM that Starship itself can be launched, left in orbit, and retrofitted into a space station module. Just purge the tanks, cut up the methane tank's dome, the common dome, and the downcomer, disconnect the engines (and return them to Earth, maybe) or maybe build smaller tanks to feed the engines for boosting the station, and finish the interior using materials shipped as the payload. IIRC one Starship would have as much volume as the entire ISS, but I can't check right now.

View on HN

One (future) Starship's payload bay alone has more volume than the ISS—that's not counting the fuel tank volume that shares a common bulkhead. The payload bay takes up just ~1/3rd of the spacecraft length.

- "The Starship spacecraft is 50.3 m (165 ft) tall and 9 m (30 ft) in diameter. It uses 6 Raptor engines, three of which are optimized for use in vacuum.[6][22] The engines produce 14,700 kN (3,300,000 lbf) of thrust.[17] The vehicle's payload bay is planned to measure 17 m (56 ft) tall and 8 m (26 ft) in diameter with an internal volume of 1,000 m3 (35,000 cu ft); slightly larger than the ISS's pressurized volume.[23]"

https://en.wikipedia.org/wiki/SpaceX_Starship#Starship_space...

I remember a common talking point for space shuttle haters (aka people who use the word "boondoggle" way too damn much) was that it's a truck and we don't need a space truck. If it was a truck what is starship?

A much much much cheaper truck.

It's a very good point! From one perspective, you can view Starship as a challenge that SpaceX can implement the STS (Shuttle) concept much better than the original STS itself. (And doing so, become cheap enough to create a high-volume orbital launch market). It's correct to be skeptical, even with SpaceX' track record. (Was it pg or whoever on HN who kept railing on that point about YC startups, that the implementation is everything—high-level ideas and concepts come cheap).

If Starship's proven a lot of things, there's still major show-stopping uncertainties remaining, like if they can execute well on the reentry heat-shield tiles. That was overwhelmingly difficult for STS. I'm not knowledgeable enough to guess if SpaceX is on a promising track with those or not—I'm looking forward to finding out what happens!

(Here's an open-ended question: if the modern SpaceX were asked to build a reusable spacecraft with (broadly) the same design choices as STS, would the outcome be closer to the dismal failure of the original, or to the success of Falcon 9? That is to say: how much of SpaceX' success do they owe to smart high-level design choices, vs. effective low-level execution?)

By same design choices do you mean that it is mandated that an essentially reusable cargo truck be mounted to the side(s) of booster(s)/tank? Because that was the shuttle's damning sin. Besides the collapse of the USSR there's a reason the Buran wasn't further pursued, and commercial space hasn't pursued a similar form factor.

It's using the same engine as the Space Shuttle (literally - it's using refurbished engines from the Space Shuttle [1]), exact same solid rock boosters with all their flaws, and so on. One difference from the Shuttle is that the SLS doesn't even have a pretext of reuse. It's a single-use only rocket.

One other big difference is that it's also obsolete before it's even complete. It can lift about 50% more than a Falcon Heavy for what will be orders of magnitude greater cost, to say nothing of comparisons to Starship. So there's no viable use case for it to ever actually be used, except for corruption. So it will be used, but it's just been a monumental waste of money that could have instead been directed towards meaningful progress in space. The entire Apollo program which spanned 11 years and started from zero, only cost $180 billion in total, inflation adjusted - less than $18 billion a year. NASA's current budget is $26 billion per year.

[1] - https://en.wikipedia.org/wiki/Space_Launch_System#Core_stage

> if the modern SpaceX were asked to build a reusable spacecraft with (broadly) the same design choices as STS, [...]

That depends on which design choices should stay the same. The boosters and the main tank were bad design choices that made reuse unattainable. With Falcon 9's and Starships reusability lessons in mind perhaps one could make the boosters just part of the main tank, all methalox burning, and have the main tank land itself.

With the booster and main tank design choices disposed of, the remaining major design choice would be that the shuttle had to be a space plane that could land like a plane. That seems like a tremendous waste of weight which would keep the new program from being broadly successful at being cheap, though it might not be fatal I suppose. But at this point Falcon 9 has a better reliability record including landings since they figured out landings than the shuttle did altogether since the STS became "mature".

At a deeper level, the shuttle failed because technical success was secondary to political success. It wasn't started because Nixon wanted cheap launch; it was started because Nixon wanted votes in California in 1972. Legislators didn't support it for technical reasons, but because it brought funding to their constituents. If ever technical decisions had to be made they had to be filtered politically, and only if they continued to meet political constraints could they be adopted. In particular, optimizing it to reduce manpower requirements was contrary to political goals.

In contrast, SpaceX is focused on technical/economic success as the first and overriding goal. As a result, they've run rings around what NASA could do. NASA is by its nature hamstrung in comparison; witness the abortion of SLS, which is showing the same pathology in even more virulent form. NASA didn't fail on the shuttle because of particular choices made, but because NASA essentially could not make good choices.

Europe saw the same problem with Ariane 6.

> At a deeper level, the shuttle failed because technical success was secondary to political success. [...]

Yes, there is also all of that.

That's another thing: neither SpaceX nor any other private company would make the same design mistakes as the space shuttle program for the simple reason that private companies want to succeed. But government agencies will gladly make the same sorts of mistakes in slightly different ways (as in the SLS case) precisely because success is not actually required and none of the people making decisions have anything to lose anyways.

And even once it was "mature" the number of near loss of craft events that happened was far too high. There were many times they landed missing too many tiles, more tiles than they thought.

> One report describes the crew as "infuriated" that Mission Control Center seemed unconcerned. When Gibson saw the damage he thought to himself, "We are going to die"; he and others did not believe that the shuttle would survive reentry. Gibson advised the crew to relax because "No use dying all tensed-up", he said, but if instruments indicated that the shuttle was disintegrating, Gibson planned to "tell mission control what I thought of their analysis" in the remaining seconds before his death.

> Mullane recalled that while filming the reentry through the upper deck's overhead windows, "I had visions of molten aluminum being smeared backwards, like rain on a windshield". Although the shuttle landed safely, "The damage was much worse than any of us had expected", he wrote. Upon landing, the magnitude of the damage to the shuttle astonished NASA; over 700 damaged tiles were noted, and one tile was missing altogether.

https://en.wikipedia.org/wiki/STS-27#Tile_damage

The Space Shuttle wasn't a boondoggle because it was a truck; it was a boondoggle because it was an incredibly expensive truck. It failed utterly at the ostensible and valid reasons it was constructed, to make access to space cheap and routine.

Where the shuttle led, cyber truck has followed.

The cyber truck is just a follower. The Roadster got there first:

https://where-is-tesla-roadster.space/live

The reason it was expensive was the military specs (be able to land back home after one orbit etc). The goal of the space shuttle was also to help drive USSR to bankruptcy and it delivered. The soviet counterpart was a huge resource drain and did exactly nothing.

The reason for military specs was the case for the Shuttle was so marginal it required fantastical projections of launch rates and launch market to make the numbers work (and all that was far too optimistic, as subsequence experience showed). This required all government launch customers to use it. NASA desperately needed the military on board; the military didn't want it and had to be strong armed into supporting it. That military requirements were then added has to be the responsibility of NASA (or of NASA's political masters) for forcing the military to be a customer.

It was designed to put classified satellites in orbit and possibly go recover them. In the long run it somewhat paid off for the repair Hubble missions.

It was also a pretty cool truck. It generated 21kW of power, using Hydrogen/Oxygen fuel cells, and captured the output water for use on missions. During most of it's life it had to send the excess water overboard, but when it came to ISS, they started bagging it up and transferring it to the station.

The power itself, the payload bay, the payload manipulation arm made this a pretty nice platform to build ISS with. Which was a natural consequence of the shuttle not being good for much else than building space stations and recovering spy satellites before the Russians did.

In terms of a natural step in our evolution towards a space faring society it was an incredible vehicle and it inspired the imagination of everyone who watched it fly.

> It was designed to put classified satellites in orbit and possibly go recover them.

It ended up useless for both roles. The military quickly moved back to expendables. Satellite recovery never made any sense as a mission for the Shuttle; satellites are generally not worth recovering at those launch costs.

> In the long run it somewhat paid off for the repair Hubble missions.

Clearly not. For the cost of the shuttle program, a large number of copies of HST could have been built and launched on expendable boosters. This is especially the case when one amortizes engineering cost of the program over multiple instances of the physical hardware.

All the other things you mention there are examples of drawing a bullseye around where the arrow landed, post hoc rationalizations. A launch vehicle that is actually cheap, like Falcon 9 or even more so (if it works) Starship would be far more inspiring; look at the palpable excitement from SpaceX's achievements. The shuttle was a Potemkin Village in its inspiration: it superficially looked cool, but objectively it was a hollow shell, leading nowhere.

The truly sad thing about the shuttle was that the US could have had cost optimized expendable launchers many decades ago, even launchers that could have evolved into something like Falcon 9 with a recoverable first stage (the Saturn 1B might have been a good jumping off point; the H-1 engine is in the same thrust class as the Merlin, it just weighs twice as much). Bastiat's "What is Seen and What is Not Seen" should be kept in mind when thinking about the program: compare it to what the alternative could have been, not to nothing at all.

As you've noticed, the truth is not popular.

> Satellite recovery never made any sense

The idea was conceived in the 1970s. Satellite downlinks weren't that capable and several keyhole satellites still used film.

> For the cost of the shuttle program, a large number of copies of HST could have been built

That's presuming we got _nothing_ else out of the shuttle.

> post hoc rationalizations.

Yes. It's amazing that a vehicle built for one purpose was capable of being put to so many others.

> US could have had cost optimized expendable launchers many decades ago

The US is not some monolithic agency. You're ignoring all the parts of reality that are inconvenient to your argument and the fact that the shuttle was a running program for 30 years.

> compare it to what the alternative could have been, not to nothing at all.

And you're comparing it to a level of perfection that was just not achievable by a public agency with funding controlled by congress. To take this perspective and call it a "boondoggle" is clearly misguided.

Anyways.. wake me up when a private company has managed to put a space station into orbit.

> The idea was conceived in the 1970s. Satellite downlinks weren't that capable and several keyhole satellites still used film.

Specifically, the mid 70s KH-9s had a re-entry vehicle magazine that carried something like 6 or 7 REVs, and when it ran out the KH-9 became useless. The idea was that the shuttle could service the KH-9s on orbit and replace the magazine.

The real reason this became useless was that around the same time CCDs were invented, and KH-11s (from 1976 - still in the design phase of the shuttle) didn't need re-entry vehicles at all. Oddly the USAF still operated KH-9s with their REV cannisters up til 1980.

(If you're wondering what happened to KH-10, that was the cancelled 'MOL' program: https://en.wikipedia.org/wiki/Manned_Orbiting_Laboratory )

CCDs weren't even needed. The Soviets had a fax machine on their lunar orbiter to send back pictures.

A freighter?

ironically those people love the agencies who forced nasa to make the shuttle extremely expensive, because the secondary purpose was to bankrupt the Russians, who were beating them in all aspects of the space race.

shuttle delivered even on that goal. akira2501 comment explain how it was awesome in the number one goal.

While the wet-tank concept has been explored since Skylab, that's a lot of work to do before your station becomes habitable - and where do you stay while doing the work?

Launching inflatable stations inside the cargo area of Starship seems to be a much better approach, as all the fitting-out can be done on Earth, so the station is immediately inhabitable.

You answered your own question in a way - use the inflatable habs while finishing the Starship itself. Then you've got both.

It's not like you can't have several Starships parked next to each other either. There are at least 3 lined up at Starbase in various stages of production.

The payload bay could have enough finished quarters and life support to support the astronauts who would do the expansion of the station into the fuel tanks. But even if that seems a bit silly you could lash two Starships together and only use their payload bays and still have way more interior space than the ISS does now and do it all for much cheaper because Starships are so cheap.

With a pair of Starships it would be tempting to put them on either end of a tether and spin them for artificial gravity. The benefits for construction and comfort may outweigh the benefits of proximity. And it may be practical to convert from one configuration to the other.

I think you're missing a lot there, like electricity, plumbing, and heat dissipation, just to name a few things.

A lot of that would be sent up in the payload bay, and the rest in another trip. The ISS wasn't built in a day, but this could be built much faster.

Reply on news.ycombinator.com