More missions, less money, higher risk: NASA's back to the '90s playbook

NASA's budget and its new administrator's statements are evoking a ghost from the agency's past: Faster, better, cheaper. The agency closed out the last century with a near decade-long experiment in doing more with less. Administrator Dan Goldin championed the philosophy as NASA faced criticism that its flagship programs, …

[Aladdin Sane]

Faster, better, cheaper.

Fitter, happier, more productive.

[lvd]

Failure rate

"Difficult to achieve without accepting a higher failure rate and simpler mission profiles the risk of more dead astronauts."

There, fixed.

[Pascal Monett]

"a higher failure rate"

Given the costs associated with any launcher these days, I would prefer that using those firecrackers result in success, not more overall failures.

[Yet Another Anonymous coward]

Re: "a higher failure rate"

Not necessarily. With cheap regular launches you build an assembly line of missions based on the same satellite bus with newer detectors rather than spending a decade building a space-grade telescope using decades obsolete 'proven' tech

[Roland6]

Re: "a higher failure rate"

If you start out with the objective of “cheap regular launches” and a shoestring budget, you will get “cheap” and a high-level of fallure. However, a better starting point is having the objective of building a bus and its associated mission assembly line, as this will permit you to better scope the problem and seek an appropriate budget. Giving the potential for those future launches to be as uneventful as getting on a bus.

[Flocke Kroes]

More samples available

Falcon and SLS are not the only modern rockets:

Space Shuttle Ares SLS was started in 1974 2004 2011, cost $211B+$9B+$31B from tax payers and had (successes/attempts) 133/135 0/0 2/2 launches starting 2022.

Falcon 1 started 2002, cost $100M (private funding) and launched 2/5 times - first success 2008.

Falcon 9 started 2004, cost Falcon 1 + $300M (private funding) and launched 632/635 (+1 rocket and payload destroyed in a static test fire) starting 2010.

Falcon Heavy started 2011ish (the idea had been considered much earlier), cost Falcon 9 + $500M (private funding) and launched 12/12 starting 2018.

The big race for old/new space was SLS verses Falcon Heavy. I wish I could find the quote again from a NASA administrator: "Falcon Heavy is a paper rocket. SLS is real." Looks like the Falcon series is a clear winner over shuttle/Ares/SLS.

Starship started 2016ish (carbon fiber) and switched to stainless steel in 2018. Cost is at least $10B so far, mostly private funding. Launches: 6/11 tests. Actual payload to orbit: perhaps late 2026?

Electron started 2006, cost $100M private funding. Launches 83/87 starting 2018.

New Glenn started 2012-2016 cost $2.5B (from Jeff Bezos). Launches: 2/3 starting 2025.

Vulcan started 2014 cost BE-4 engine + centaur heritage + $1.2B from tax payers + probably something like $1B private funding. Launches: 4/4 starting 2024 (with SRB issues).

Firefly Alpha started 2014, cost $100M launched 4/7 starting 2023.

Ariane 6 started 2014, cost €3.7B and launched 6/7 starting 2024.

Landspace Zhuque started 2015, cost ? launched 4/6 starting 2023.

Old space has consistent success at a high price. New space results are more mixed, but much cheaper. New rockets take about 10 years with either development model whether starting from scrattch or with an experienced company building off heritage products. I want to finish this post before dinner so I will not compare Orion/K-1/Starliner/Dragon/Dreamchaser.

All we have so far are budget proposals. After being hammered by lobbyists the actual budget usually ends up much closer to last year's than the one from the White House or the first draft from the senate. I think we will see a move away from the big really impressive (late + expensive) missions like JWST. I am expecting cheaper and shorter missions to nearer destinations (the Moon). I expect that these missions will fail on the first attempts. I hope they will try again until they succeed then crank out repeats missions to different part of the Moon. I hope this will create experience and standard components that will one day make impressive missions possible on time and within budget. I know - I am being very optimistic.

[frankvw]

Re: More samples available

"I think we will see a move away from the big really impressive (late + expensive) missions like JWST..."

That would be a shame. You simply cannot develop, build, launch, deploy and commission something like the JWST quickly and on a shoestring budget. And if you want the sort of scientific data that the JWST delivers, a half-baked, scaled-down "good enough" substitute won't do. So a move away from late+expensive projects like JWST means a move away from the quality of results such projects deliver.

[Flocke Kroes]

Re: beyond JWST

You are right, discoveries mostly come from from the biggest telescopes. I hope that the move away will be temporary and there will be a move back later - with experience, infrastructure and mass produced standard components to put around really big mirrors and sensors.

[Spazturtle]

Re: More samples available

The core issue with the JWST was that it was designed to use technology that didn't exist with the expectation that technology would continue to progress at the rate it was in the 1990s. But technological progress slowed rapidly from the late 1990s to the early 2000s.

You could now build something similar to JWST with COTS parts for much cheaper and in a shorter time frame. Being innovative and at the cutting edge is not worth it if it means the project gets so delayed that your design is no longer cutting edge.

By the time JWST launched it's expensive bespoke main imaging sensor was outdated and the manufacturer was selling an updated, more sensitive and higher resolution version commercially. And if you designed a new telescope to launch using Starship then you could forgo the expensive, complicated and high risk origami system.

[Aladdin Sane]

Re: More samples available

Technological progress didn't slow, it simply went in directions they hadn't considered.

[Patrician]

Re: More samples available

"And if you designed a new telescope to launch using Starship then" you'll be waiting a long, long time based on current Starship progress.

[Aladdin Sane]

Re: More samples available

You can launch it on Starship as soon as you want, it'll just come back to earth much quicker and in far more pieces than you want.

[Spherical Cow]

Re: More samples available

Starship's development phase is almost complete, and extra manufacturing facilities are being built. Expect to see development completed this year, several commercial launches next year, and LOTS of commercial launches the year after.

[IGotOut]

Sigh...

"as NASA faced criticism that its flagship programs....demanded years of development and billion of dollars. "

Or one day of conflict in Iran

Just found this, not sure how accurate it is, but fucking hell!

https://militaryspend.org/us-iran-war

[MachDiamond]

Re: Sigh...

"Just found this, not sure how accurate it is, but fucking hell!"

Just the amount of money that the US WAR department can't account for would nicely fund NASA for over a decade. Technologies developed by NASA are mostly made public where commercial entities can license its use. Military tech, not so much. The ROI for taxpayers hugely favors spending money on NASA and space programs. Paying commercial companies to do space development locks up the IP which makes no sense if they are being paid with taxpayer money.

A nice side benefit is that images/video created by NASA are public domain. Some images one might see online may have been created by a private entity, so care should be taken, but there are lots of images available from NASA.

[Roland6]

Re: Sigh...

Another side benefit of NASA is that the IP is owned by the (US) public and not the property of corporate entities…

[Jou (Mxyzptlk)]

Won't change until Apollo 1 incident is repeates...

Nearly 60 years ago. Challenger happened due to political pressure as well. A lot of people died that way.

[Flocke Kroes]

Re: Won't change until Apollo 1 incident is repeates...

The current budget gives Artemis 14% more than NASA requested - probably subject to the inevitable expense causing delay when Trump tries to negotiate changes. There is a fair chance budget will not be the problem for manned spaceflight - especially as the HLSs are firm fixed price. Its the robots that are currently set to experience budget cuts.

The interesting bit will be the Artemis 4 launch date. Trump's target is before the end of his second term. He is working hard on being sufficiently unpopular to get kicked out early. He certainly intends to be president for life but he could die of old age or assassination. This makes the target date as unpredictable as when the hardware will actually be ready.

Orion is still a small cause for concern. The Orion for Artemis 3 will have a different heat shield that the ones for A2 and A1. It will not be stressed as hard as A4 because A3 is only returning LEO. It would be nice to test the heat shield with an uncrewed mission. That is impossible because SLS launches are so expensive. Last time fixing the heat shield got a hefty budget for simulations and Earth based experiments - pretty much anything is cheaper than an SLS launch. I have confidence that old space lobbyists can get the required time and budget if Orion needs more work. Their ability is this regard is outstanding and has decades of demonstrated reliability.

The contract for the HLSs requires a successful uncrewed demonstration landing before trying a complete mission with crew. NASA renegotiated the SpaceX contract to include testing ascent. I have a strong lack of confidence in either vehicle working perfectly on the first attempt. Putting crew on the demonstration mission would be exceptionally courageous. NASA successfully resisted pressure to put crew on Artemis I. I think they would be even more terrified of putting crew on a demo HLS. Trump would have to replace half of NASA to advance the schedule to match Apollo 1/Columbia/Challenger safety levels.

Move fast and break things will apply only to robotic missions.

[Irongut]

Re: Won't change until Apollo 1 incident is repeates...

> Move fast and break things will apply only to robotic missions.

History suggests this is not correct.

[Roland6]

Re: Won't change until Apollo 1 incident is repeates...

People forget just how important the words "and return him safely to the Earth" were to early US thinking about manned space flight. It does seem this concept is being quietly watered down with the emphasis on more with less…

[MachDiamond]

I beg to differ

"SpaceX, for example, has demonstrated that launches can be done relatively cheaply, and its rapid-iteration model has produced real results. But it has also produced real failures: Starship, despite reaching its third major version, is only now approaching the reliability threshold for uncrewed orbital flight."

SpaceX has been holding funding rounds a couple of times per year to keep the checking account topped up. None of that money from one Elon Musk. That market being tapped out with no real ROI, SpaceX has to go public. SpaceX also charges very close to what the Russians were charging per seat for crew rides to ISS. Cargo is cheaper as there are competitors for that business. ISS is coming to the end of it's life which axes that revenue stream from SpaceX

Starship is starting over with V3 to launch in mid-May 2026 as a mostly empty shell, again. They have yet to try for orbit and catching both stages. They haven't delivered a HLS mockup to NASA, that I'm aware of. Not even realistic CGI. There's a tanker model to build, an orbital depot, a cargo version (that hadn't been discussed previously) and real hardware for taking crew down from lunar orbit and back. There has been work on the Pez dispenser and trialing launches of Starlink sats. The NASA administrator is talking about using Artemis 3 to test docking of Blue Origin and SpaceX lunar landers in Earth orbit in 2027. Everybody better get a taste for triple espresso's and no holidays to meet that timeline.

[Flocke Kroes]

Re: I beg to differ

Most of those "funding rounds" are secondary sales so SpaceX gets nothing from them. Dragon is a small potato. The money has been coming from Starlink. The IPO is to pay for Musk using SpaceX to buy xAI for a ridiculously high price. Its a pattern that goes back at least as far as Tesla buying SolarCity: use a successful company to buy one of his failures. Effectively outside investors pay to give Musk a higher proportion of the shares in the successful company. So far they have fallen for it each time.

[Anonymous Coward]

Please send Trump and Elon to Mars soon

Bye!

[Bebu sa Ware]

Faster, better, cheaper is back

"history suggests you can't get all three at the same time"

The answer to question of whether you can ever get all three, if not blowin' in the wind, is liberally spread over the landscape.

[vtcodger]

Re: Faster, better, cheaper is back

Can you get all three (Cheaper,Better,Faster)? Probably. Occasionally. Back in my working days, it crossed my mind that just as some projects seem doomed to fail, some projects seem doomed to succeed. No matter how weirdly run, everything comes together. Things just work. It's almost as though management has a clue. The problem is that when that happens, the managers are regarded as geniuses. And their toolkit is viewed as having magic powers. For a while everyone MUST follow the path of the wizards. Oddly, the magic doesn't seem to work for anyone else. Neither does it work (usually) for the geniuses in their future efforts.

[MachDiamond]

Re: Faster, better, cheaper is back

"It's almost as though management has a clue. "

That would be novel.

It's more that there was a good crew doing the work and the stars aligned so things came in on time and the machinists got everything made to tolerance in the correct material. Time to do things twice is a minimum budget. The first one always takes ages while the second one bypasses all of the mistakes made on #1. It may not be perfect, but it can be screwed up must faster with more insight.