NASA's SLS rocket's Mobile Launcher-2 increase from $383M to $2.7B
twitter.comHow is it that government contractors can't build this thing on even a 9-year engineering schedule (July 2019–September 2028) when SpaceX is iterating almost exactly the same thing, on a timescale of months? They've already launched 4 SLS-sized vehicles, and have very visibly been redesigning the ground-support equipment each time around. The next launch for instance (next month!) debuts mechanical arms attached to the launch tower, which will attempt to capture the flying booster on its return.
What exactly is it Bechtel and friends struggling with? The OIG document doesn't answer this in a way I can understand.
NASA is a political jobs program from a funding perspective. There isn't any way around that.
This creates uncertainties, distractions and perverse incentives all around:
0. Everything is dramatically complicated by the artificial divisions of work across states needed to get enough political support. Grossly inefficient from the start, and inflexible into perpetuity.
1. Initial schedules are incentivized to be unrealistic, and not reflective of hard analysis/planning, in order to get funding. Easier to deal with the real details and expand schedules later, even though nobody wants that to be the process.
2. Essentially a cost plus (or cost even) program. There is no hard limit on cost, since the further along a project is, the easier it is to raise its budget, especially along with schedule delays.
3. Creativity flows to the most basic constraint. The most basic constraint is political will to fund, so top level attention goes to managing/manipulating up.
When NASA's hard constraint was time (getting to the moon first), creativity went there. They were never cheap, but they did operate with incredible speed.
Whereas any private company's basic constraints are relative profitability in terms of net return in a given time frame, so creativity goes to both speed and cost.
4. Boeing and other traditional large NASA contractors operated for decades under NASA's politically driven, cost plus umbrella, so continue to suffer from, and reinforce, NASA's problems.
This culture is hard to break, even after switching to a flat cost (still time flexible) regime. See recent Boeing troubles. Not just throw away rockets, but a much delayed, over budget, wonky one-way manned spaceship!! Ouch!
> 2. Essentially a cost plus (or cost even) program.
This is really the only problem.
Because the government was the only buyer - no one in their right mind would supply unless a ridiculously cushy contract is in place.
You're going to massively overpay if you're the only buyer in a new industry that requires insane CapEx and is high risk.
That’s part of the plan though. With high risk nascent industries, the govt is often the only customer who can bear that risk. It was the same dynamic that ushered in aerospace; it was mainly just a hobby industry until the Army offered a lucrative contract.
>NASA is a political jobs program
This is the truth, and while you could frame it as a negative, I see it as a huge positive.
Because NASA has been subsidizing space for decades the US has a tremendous among of trained workers for aero-space. It was exactly that workforce that venture-capital based space industry was able to source talent from. I doubt spaceX would have been possible without NASA dumping billions into boondoggles. Now it is self-reinforcing. NASA trained talent for/funds industry, industry now trains talent for/supplies NASA. Space economy the easy way, just costs a few hundred billion bootstrap it.
> This is the truth, and while you could frame it as a negative, I see it as a huge positive.
I see a clear split.
When NASA does something brand new, where private industry doesn't have incentives, it is a jobs program doing pioneering science, producing unique scientific and technological progress.
Since those projects would not get done otherwise, the inefficiencies are not really inefficiencies. Just cost of project.
But when NASA does something industry has found incentives for, the result is massive money-wasting redundant lower-quality, economically deadend work. The SLS is a monstrous parasite, tragically sucking up NASA/tax-payer resources. The only "purpose" for each build, launch & discard, vs. buying an economy class Starship ticket, is to "justify" the cost of doing so!
So I applaud NASA's manned missions in the past.
And I applaud NASA's current unmanned missions exploring our solar system and universe. And asteroid deflection missions. And human habitability research.
But manned transit, and near Earth resource missions, are best left to the growing list of companies that are funded, incentivized, better managed, and single mindedly optimizing those activities so well that they not only pay for themselves, but grow unbounded.
> 3. Creativity flows to the most basic constraint. The most basic constraint is political will to fund, so top level attention goes to managing/manipulating up.
Can you or somebody elaborate on this? Is there an idea or a terminology or something I can read more about this? I'm very interested in the idea that you can pay attention to where the "smart kids" are going, to help identify where the most basic constraint is. I'm also interested in what the word "basic" means here. I'm familiar with the theory of constraints and the idea of bottlenecks. But that doesn't seem like what this is getting after. Thanks!
> They were never cheap, but they did operate with incredible speed at that time.
Actually, they were good, fast, and pretty cheap. The Saturn V cost per kg was much something around $5000/kg (compared to Delta IV Heavy at around $12000 kg), and could send a ton more into orbit than anything until Starship. If we'd continued down that development line instead of the insanity of the Shuttle, we'd really be somewhere today.
They also inherited a lot of free R&D from the Nazi V2 rocket program via Wernher Von Braun and several of his collaborators thanks to Operation Paperclip…
Yes, so?
It’s not like that knowledge was forgotten and rediscovered post 1972.
The US government is a jobs program. When you have been the dominant superpower for over 30 years, there is no reason you should be effective or efficient at anything. If nobody is capable of challenging you, private interests tend to override public interests.
And when a real challenger arises, you just have to hope that you can discard the old establishment and replace it with something capable of delivering before it's too late.
This is just my opinion, but I think a lot of it has to do with requirements.
NASA is first and foremost about prestige (even above science IMO. This is also - again my opinion - why NASA hasn't been back to the moon in 50 years. There's little prestige to be gained for an expensive operation). With prestige as the backdrop, it makes them very risk adverse. There are layers and layers of requirements made to reduce risk. To their credit, SpaceX is willing to fail more than NASA and it allows them to iterate much faster.
This is also my opinion part of the reasons why NASA wanted the CCP to begin with: it allows them to skirt many of the requirements. The somewhat ironic part is that NASA pathways to alleviate those requirements, but it's somewhat common that project/programs want to go through the waiver process and formerly accept that risk.
There is a fairly well-known dynamic in govt contracting where a contractor low-bids to get the contract and then makes money on change-orders. If they were to accurately bid on all the NASA requirements, they risk their bid being so high they would never be awarded the contract. This is a similar dynamic to the traditional cost-plus contracting paradigm.
There is also some irony in that SpaceX may be the victims of the same dynamic. As failure occur, they may layer on more requirements that ultimately slow the process down (see the supplier quality issues related to an F9 strut failure). Given enough time, there is a risk that added requirements turn them into the dinosaurs they are replacing.
Also, as others have brought up, NASA has additional requirements they must manage (e.g., political risk) that SpaceX has less exposure to.
> NASA has additional requirements they must manage (e.g., political risk) that SpaceX has less exposure to
This doesn't appear to be the issue with ML-2, whose problems are closer to Boeing's subcontracting addiction than e.g. the F-35's sticker-collection approach to manufacturing.
I’m sorry, I don’t follow. Can you elaborate? Maybe I’m misunderstanding, but I think sub-contacting is heavily related to political risk mitigation, as it’s a method of spreading project money across many political districts.
IMO NASA has done the same with their selection of major center locations. It doesn’t make a lot of logistics sense to have your biggest centers in CA, OH, VA, and TX but it makes a lot of political sense if your goal is to avoid large programs being mothballed.
> sub-contacting is heavily related to political risk mitigation, as it’s a method of spreading project money across many political districts
Good point, I'm unclear whether Bechtel was awarded the contract by NASA/Congress after selecting its subcontractors or before. (And if having a roster of subcontractors was an implicit ask from NASA or a decision made by Bechtel to future-proof their contract.)
We see something similar with Boeing, and while that was also partly politically motivated, there was also the financial desire to be asset light. The OIG's story about "one subcontractor, [to] which Bechtel allocated approximately 46 percent of the fabrication work to" selling off "all of its shop space to a non-NASA customer because Bechtel's steel fabrication plan lacked a signed contract with the subcontractor" forcing Bechtel "to find another subcontractor with available shop space" sounds like incompetence, not savvy politics [1].
[1] https://oig.nasa.gov/wp-content/uploads/2024/08/ig-24-016.pd... page 19
SLS stands for Senate Launch System. It's a pork distribution program.
Major beneficiaries are the states of Florida, Mississippi, and Alabama (Huntsville) but there are beneficiaries all over the place. Classical DoD/NASA contractors know to locate facilities all over the country to maximize political capital via connection to local jobs in many jurisdictions.
Well, they do a lot of fundamental try/test that others use then, including SpaceX. Didn't want to defend NASAs schedules, of course, but still they are often the first pioneers in many areas. And that plays not really well with public money, unlike private startups that can do whatever.
NASA still does great stuff but SLS specifically is not exactly a pioneering program. It's a disposable rocket using Space Shuttle engines, attempting to go to the Moon again.
Meanwhile SpaceX got reusable first stages into production, and recently got the world's largest-ever rocket to orbit and attempted to land the second stage. And it's all powered by newly designed engines that significantly advance the state of the art.
There are _a ton_ of innovations in SLS, especially in the space ship (payload). Engines is just a small piece of the program, not really important one. Over the years a lot of theory improved, a lot of new technologies developed (e.g. materials, IT) and SLS tests them for space. Kinda "deploy in production".
Could you be more specific? Because as a launch rocket, it's so absurdly expensive that it will never be useful other than for low-volume national vanity projects. What technologies has it developed that might be usable on other platforms?
> What exactly is it Bechtel and friends struggling with?
Some of it is reasonable. Covid threw a spanner in schedules and increased costs due to a rise in price levels. Some is from design changes due to new information, e.g. "during launch, the SLS generates exhaust blast plume pressure, random vibration, vibration from acoustics, and heat. These loads can cause damage to the launch vehicle, payload, launch pad, and surrounding structures. After the Artemis I launch, NASA found higher-than-expected thermal, acoustic, and blast loads to the ML-1. ML-2 project management is assessing these lessons learned from Artemis I and anticipates the ML-2 structure will require some additional strengthening to withstand the predicted loads" [1].
A lot of it, however, is laughably-incompetent cost-plus subcontracting nonsense: "While Bechtel revised its IFCs in response to the RFIs, according to ML-2 project management, the company did not allot sufficient time in its schedule to do so, and this iterative process resulted in cost increases and schedule delays. In mid-2023, ML-2 project management found the delays in steel fabrication and delivery resulted in a 3-month schedule slip, leaving Bechtel with no additional schedule reserve to meet the May 2026 contract end date.
...
In mid-2021, ML-2 project management noted that Bechtel’s 'interactions and business relationship with the steel fabricator deteriorated to the point of dysfunction,' resulting in unresolved fabrication issues that impacted the ML-2 project’s critical path. Bechtel’s lack of awareness and oversight of critical second-tier subcontractors responsible for steel fabrication contributed to a delayed construction start date. One subcontractor, which Bechtel allocated approximately 46 percent of the fabrication work to, sold all of its shop space to a non-NASA customer because Bechtel's steel fabrication plan lacked a signed contract with the subcontractor. As a result, Bechtel attempted to find another subcontractor with available shop space but was unable to do so in a timely manner. The delay in the steel fabrication process continued to impact the ML-2 project’s schedule" (pp 18).
[1] https://oig.nasa.gov/wp-content/uploads/2024/08/ig-24-016.pd... page 20
SpaceX is like Google, whereas NASA is worse than Oracle because it's the federal government AND a highly regulated industry involving human safety. This is in so many ways, hiring to engineering processes.
> whereas NASA is worse than Oracle
A lot of Artemis is dictated to NASA by the Congress. Look at NASA's deep science missions, e.g. landers, probes and telescopes. They're ridiculously advanced, creative and cost effective for the amount of science they deliver.
all space projects are R and D projects....how do I know? jet engine technology took 50 years before we could go years without a test fighter pilot dying....
fun fact...its part of the legal boilerplate astronauts sign...
Sure, but the amount of R&D varies.
It's frequently called the "Shuttle Launch System" (https://en.wikipedia.org/wiki/Shuttle-derived_vehicle) because "it uses existing Shuttle tech!" was one of the big selling points.
Keep in mind that Starship has never made it to orbit.
Yes, but it *launched* four times, and the launch tower verifiably works, and it didn't take a decade to build. That's the subject of this question, isn't it? The $2.7 billion NASA wants to spend on a steel-frame launch tower.
In that case there is no SpaceX equivalent. Is SpaceX's tower used for transporting stacked human-rated rockets from hangar to pad?
This is very technically correct but also misleading.
In the last test flight, Starship reached a velocity of 7.3km/s. Orbital velocity is 7.8km/s.
Why didn't they make it to 7.8km/s? Because they choose not to. Starship had the fuel, control and capability to burn a few more seconds for another 500m/s.
It also (barely) survived re-entry and achieved a fully controlled, powered, soft splashdown.
Because the mission profile has been intentionally sub orbital to avoid any possibility of uncontrolled reentry. There would be minimal change in the burn for the last 2 flights to place it in orbit.
Musk level leadership, delegation, and drive is once in a century. It’s hard to blame contractors for basically acting like every other comfortably large organization.
OAG report on Launcher 2: https://www.oversight.gov/report/NASA/NASA%E2%80%99s-Managem...
The contractor is Bechtel.
Note, the Senate Launch System is already a huge success: money was transferred from the taxpayer to the contractors as a "jobs program". It doesn't matter if anything flies.
That't s a good link, thank you.
> NASA intends to keep Bechtel accountable to the cost and schedule agreed to in December 2023.
It's mind boggling how they heck is Bechtel qualified to handle this project? Bechtel is at the center of https://en.wikipedia.org/wiki/Cochabamba_Water_War / https://www.sfgate.com/news/article/bechtel-battles-against-.... Talk about an agency that supposedly cares to look progressive and caring for the environment. Then they pick Bechtel as their contractor for the SLS.
> We also found that Bechtel’s performance drove the significant cost increases and schedule delays to the design and development of the ML-2. The current contract value of $1.1 billion includes $594 million of Bechtel overruns
Shouldn't somebody there fire Bechtel. I guess that significantly reduces your chance of retiring at Bechtel after leaving your NASA job...
> Shouldn't somebody there fire Bechtel
This means cancelling Artemis II.
I was exaggerating. It's, of course, too late for that. The OIG recommended "learning a lesson" at least:
> (1) ensure lessons learned from the ML-2’s acquisition, contract, and project management are codified to inform future development efforts a
And it will be 3 years late!
I honestly would love to know how much money is 'wasted' by NASA because of these companies that seem to, over promise and then need a lot more money.
And I don't say wasted because I think that the money spent at NASA is not worthwhile. But how much else could be done if that money could be spent elsewhere or at the very least be properly estimated in the beginning so it could have been planned for. That money is (I assume) going to come at the cost of something else.
I really don't love the idea of SpaceX not having a serious competition. But... they kinda don't right now anyways it seems. I really hope we have another company step up to be anywhere near what they are doing.
Stoke Space seems like it has potential. Same engineering philosophy as SpaceX, really innovative design, going for 100% reusability from the start. Everyday Astronaut visited them last year:
Bold design. Re-usable actively-cooled cryogenic second stage is wild.
Say what we will about Elon, but him gambling his PayPal fortune was the best thing that happened to the US space program (and the global space race as whole!)
100% - 1 rocket failure away from financial ruin!
It’s crazy how different the world would be right now. Ukrainian generals admitted they would have been defeated quickly without Starlink.
https://www.theguardian.com/technology/2023/sep/07/elon-musk...
> Musk told engineers to turn off communications network to hobble Ukraine drone attack on Russian warships
Did you see the amendment at the top?
"Update: on 9 September 2023, Walter Isaacson said his biography’s claim about Starlink and Crimea was based on “mistaken” information [see footnote]"
Here's a more accurate summary: https://en.wikipedia.org/wiki/Starlink_in_the_Russo-Ukrainia...
Having praised Elon, I will now additionally add that he is pretty much a mouthpiece for rightwing Russian propaganda despite Starlink being vital to Ukrainian war efforts.
You've truly achieved wealth when even people you don't like and don't like you in return depend on your shit to stay alive!
Context for others:
Both sides in the Ukraine War are using "irregular" Starlink dishes.
* Russia is reportedly buying dishes (and the service attached to the dishes) in the Middle East.
* Ukraine has its own fleet of dishes (being paid for by the US, after Musk initially provided free service early in the war after a Ukrainian request), but also many individual dishes that were donated, and being paid for, by private individuals outside Ukraine.
US law prohibits Russia from using Starlink. The problem is, how to stop Russia from doing so? A simple location-based ban won't work, because the front line is constantly shifting. Whitelisting only Ukraine's own dishes to work within Ukrainian territory might work, but 1) what about dishes that get captured by Russia? 2) As noted, what about all the privately paid-for dishes?
Another way to think about this is that this demonstrates just how lifesaving for Ukraine Starlink has become in the past two years. Ukraine could ask Starlink to disable all dishes within its territory. On the contrary, it has decided that the benefits of Starlink to Ukrainians outweighs Russians also benefiting from it.
von Tiesenhausen's Law of Program Management- To get an accurate estimate of final program requirements, multiply the initial time estimates by pi, and slide the decimal point on the cost estimates one place to the right.
For perspective 2.7B is 0.04% of the entire US budget for 2024. This is a rocket that goes to the moon. If your reaction is "lord help us" then perhaps space exploration just isn't in your blood.
This isn't for the rocket, just the structure that it's connected to before launch. Tens of billions of dollars have already been wasted on the rocket. I say "wasted" because it's already essentially obsoleted by the Falcon Heavy which has been ready to go for years, is reliable, and launches at a tiny fraction of the cost. And by the time Starship comes into play you're talking about a Ford Pinto being sold at Lambo Veneno Roadster prices.
Just to add on, if you don't mind: the SLS uses four RS-25 engines, the exact same engines that were used on the Space Shuttle (admittedly, there is a plan to improve them later), as well as solid rocket engines that are derived from the Shuttle SRBs.
I actually do think that more than one US-based provider of space launch systems is something important, but it's wild to me that mature US launch systems can basically be cleanly divided into "companies trying to sell the same technology to the government for the fifth time over fifty years" and "SpaceX."
I'd like to expand on this and clarify for anyone reading this that it's not just using the same type of engines as the shuttle. They are literally the same reusable RS-25 engines taken off the shuttles over a decade ago and used for one final expendable launch. Likewise the SRB segments are reused from shuttle launches, but reconfigured and expended for one more launch on SLS. The $32 Billion spent so far on SLS have been to rearrange pieces from the shuttle parts bin.
There are further contracts to make new RS-25 engines and SRBs optimized for expendability once they run out of old shuttle equipment.
At this point would it have been cheaper to make Shuttle 2.0 with all the various flaws fixed lol
Falcon Heavy, fully expendable, can send about 15,190kg to TLI (Trans Lunar Injection). The Saturn V sent about 48,600kg to TLI.
The current SLS can send about 27,000kg to TLI, and block 2 is expected to send between 43,000kg and 47,000kg to TLI.
Could the Falcon Heavy be used on a multi-part moon mission? No doubt, but it's not in the same weight class as SLS.
The Falcon Heavy can ship can ship 16.8k kg to Mars [1] and that's at a higher delta V distance/cost than TLI. In general the SLS can carry around 50% more than a Falcon Heavy, but the Falcon Heavy (fully expended) costs 5-10% as much.
Thanks for the clarification. I dug around for numbers but didn't look in the most obvious place!
And yes, the cost difference is absolutely staggering.
I’m honestly shocked more “assemble in orbit” type mission structures haven’t been proposed. Falcon Heavy is just so convenient yet ignored.
> shocked more “assemble in orbit” type mission structures haven’t been proposed
It's unprecedented. A large part of Startship's technical risk in the Artemis programme is in-orbit refuelling.
That said, if IFT-5 or 6 can demonstrated recapture and an in-orbit fuelling attempt is made in 2025 that doesn't end catastrophically, I think the centre of mass of technical risk for Artemis shifts away from the novel fucking launch vehicle system that SpaceX has been tasked with building from the ground up to checks notes a rocket based on refurbished Space Shuttle engines and double checks this glorified launch tower.
Falcon Heavy has the thrust and the delta v, but it doesn’t have the fairing volume. Basically anything interesting that fits can be launched by smaller vehicles.
Starship will be a game changer in the most literal sense here as long as they figure out payload doors…
It's not for the rocket.
A moon rocket for 2.7B would be a steal. This is just for a mostly static steel tower (on a mobile carrier?) that holds the rocket up and provides some piping etc to the rocket before launch.
The mobile carrier is not included in the price AFAIK, but not sure.
It's tower on a large steel platform but it also includes significant GSE (ground support equipment) - cabling and cryogenic plumbing to connect to the rocket to let it be fueled and readied for launch, all connected via mechanical arms to pull the quick disconnects away from the rocket, in some cases into armored boxes to protect against the rocket exhaust. Plus a crew access arm to let astronauts board, elevators so they don't have to climb stairs in their space suits, etc., etc.,
It is carried around by one of the two existing crawler-transporters, which moves it between the VAB where the rocket is built and the launch pad.
What, this isn't even an new crawler/transporter? It's just the gantry and plumbing?
No need for a new one; the two they built for the Apollo program still work just fine.
Note that the crawler-transporter doesn't hang around at the pad during the launch - after setting the mobile launcher down at the pad, it drives away back towards the VAB until it's needed next.
During the Apollo era, it also transported the Mobile Service Structure - a tower with work platforms that allowed access to much of the spacecraft on top of the rocket.
Artemis has cost nearly $100B so far, and with increases like this, that seems unlikely to slow down.
If Musk wasn't going crazy, it would make sense to kill the SLS.
Shotwell seems like a solid CEO though?
Edit: she’s president and COO apparently. Musk is CEO, nm.
LOL everyone is thinking she's CEO:
https://news.ycombinator.com/item?id=41230320 also confuses the name/gender, Glenn vs. Gwynne
Do you think Musk's antics are having a notable negative effect on SpaceX? There's no doubt that's the case with Tesla.
> Do you think Musk's antics are having a notable negative effect on SpaceX?
Absolutely, politically. There is an alternate world in which Musk retained his bipartisan credentials where SpaceX is being handed the Artemis programme.
This is why Shotwell mostly runs the show at SpaceX. Unlike Tesla, SpaceX runs like well oiled machine and is kicking butt.
> There's no doubt that's the case with Tesla.
Um, yes.[1]
[1] https://jalopnik.com/elon-musk-is-turning-off-would-be-tesla...
I think we're in agreement about Tesla. I believe that Musk's bullshit has greatly harmed that company. At the same time, I don't think it's hurt SpaceX very much.
Fair point. Space exploration takes a lot of money. But even that has a limit. $2.7b is like a looooot.