NASA's Faith-Based Approach For Future SLS/Orion Budgets
A new, independent review of the Orion spacecraft is pretty damning, Ars Technica
“Despite these concerns, NASA is pressing ahead with an effort to try and accelerate development of Orion to enable an August 2021 launch of Exploration Mission-2. Yet the GAO found this scenario improbable. “To stay on the aggressive internal schedule, the agency is counting on receiving higher appropriated funds than what it plans to request, which may not be realistic in a constrained budget environment,” the report states. There is low confidence – 40 percent – in NASA making the 2021 launch date, and the GAO believes this may not be a “beneficial strategy” for Orion in the long term.”
– Double GAO Reports: SLS and Orion Cost and Risk Estimates Are Still Unreliable, earlier post
– Earlier posts on SLS and Orion
Lots of good comments on the Ars Technica article. I agree that the program is working as intended – to funnel jobs and dollars into specific districts, centers, and contractors. The deferred work ethos was standard operating procedure when I was active. “Sending regrets” was the accepted term when a budget cut came down. Once a budget was cut in a particular area, it would get cut there again and again. Pretty soon, the effort would be so far behind the state-of-the-art that it could never catch up at which point the technical staff would wear the stigma of failure. I don’t miss it.
Commercial Crew supporters better get their “congressional outreach and education” shoes on. This is how they will work to reduce funding for CC or how I would try to do it anyways.
Even with Team Alabama and Texas pushing for reductions in funding I am not sure if this stops CC at this point. We may be close enough to where the billionaires finish the job if funding gets pulled.
Congressional staffer friends – please remember: government funding keeps a more aggressive schedule with CC. The fastest way to return Americans to space is with CC. Even if you could go full Apollo crash scenario (we cannot) CC is still faster and just as safe.
CC is the better option.
The lengths to which GAO avoids saying ‘face palm’ are extraordinary.
No, it’s just reading a different language. When you literally translate the bureaucratic language into plain English, the face palm or face plant words aren’t there. But is you know the language, the implication and the subtext are there.
The GAO report, as one might expect, focuses primarily on whether the program can meet its stated cost and schedule. The question of whether the mission and strategy are appropriate, and whether the cost is appropriate for the chosen mission, are not considered.
Probably those strategic and policy issues would be considered to fall outside the GAO’s area of competence anyway.
The GAO addresses specific questions put by Congress. Should Congressman X ask the appropriate question, GAO is fully capable of answering.
GAO generally excels but can’t answer questions like ‘should NASA engage in #Mars or #Luna’. On the other hand they are very good at defining parameters and discussing the costs of each with in those parameters.
If I missed it sorry- but do we know who/which congress folks requested this report? It would be realliy interesting to see that info.
Orion was the safe, simple, soon project that was going to fly by 2011, or later it would fly by 2015 and there would be no need for Dragon or CST100 according to Lockheed, or during Augustine it was likely to fly in 2017 or 2018 (according to Sally Ride-Mark Geyer was claiming 2014), and now they are looking at 2024 or 2025. I cannot even imagine how you can design and build something this slowly. I could probably build this thing at home and it would have been done years ago.
Its not like they are building and testing lots of prototypes. They have a couple flight units in work. And remember this is supposed to go to Mars? except it has a heat shield, as the report points out, that is suitable for no further than the moon. So they need to put in a major redesign before the 2nd or 3rd flight. Musk will probably have RedDragon on Mars five years before this wasted effort even flies. And Musk is doing it for under a half billion$$???
Best comment on the Ars Technical article is that its a very efficient delivery system for sending $16billion to Lockheed and getting little in return.
Today we have no Shuttle, no Constellation, no Orion. More significant than any of these there really is no longer a vision-we have no idea where we will go, or why. Never mind no plan; we don’t even have any idea of the goal. Look at Mars for a couple hours as an Orion flys by? That is what this article implies is our long term goal.
“Orion was the safe, simple, soon project that was going to fly by 2011”
No, that was Ares I. “Safe, Simple, Soon” was ATK’s advertising slogan. Nothing to do with Orion.
What was going to be on top of Ares I?
Irrelevant. It was ATK, not Lockheed preaching “Safe, Simple, Soon”. ATK was only building Ares I. They weren’t designing or building Orion. And I don’t recall anyone ever saying Orion was going to be simple or soon.
The CEV. Later called Orion, which was the precursor to, but not the same as, the Orion MPCV spacecraft that is being built now.
Unfortunately, I can easily imagine how developing SLS could take so long and cost so much. There are easily tens of thousands of things which can go wrong with a launch vehicle. Put yourself in the place of a project manager. If someone identifies a potential risk, even an unlikely one, and says, “out of your multi-billion dollar budget, shouldn’t you find a million dollars to avoid this risk? After all, astronauts’ lives might depend on it?” What would you do? Take the risk of explaining to a Congressional committee why your bad call killed astronauts, or find an extra million and a few more months of development time to eliminate an unlikely failure mode? Then take that, and multiply it by all of the possible, but unlikely failure modes. That can add up to billions and years.
Balancing risk and efficiency isn’t easy. If you are off in favour of reducing risk, the logic of your position can push you further off. (It’s now a ten billion dollar project, not a five billion dollar one, so you need to be even more concerned about risk reduction.) I’m I can see that sliding into the current state of SLS, even without the political issues.
Most of my comments were with respect to Orion. However even in the case of SLS this was supposed to be a simple offshoot of the Shuttle. Tankage based on Shuttle’s. Engines same as Shuttle’s. SRB’s based on Shuttle though extended. So I have similar questions about a 20 year development period and in SLS’ case, looks like closer to $$30billion
“supposed to be a simple offshoot of the Shuttle”
I think much of the suppliers of various parts and systems for Shuttle were dismantled. I recall Wayne Hale wrote about “horse left the barn” regarding people wanting to keep Shuttle flying after it was announced in 2004 it will stop flying in 2010. Hale mentioned there are several “mom and pop” shops that made miscellaneous components for STS and these all closed shop. So the big stuff kept open but all those smaller parts suppliers disappeared. I feel like we are like Soviet Russia in the early days of the space race. US beat them to the Moon and many other things because we had an “Ace Hardware store at every corner” and they did not. Now much of those hardware stores no longer exist. Perhaps Musk built up “hardware stores” (i.e. vertical management or similar term), or maybe because lower level people at SpaceX don’t have to deal with horribly bureaucratic procurement processes.
Farming out the hinge plate to this mom and pop shop in this political district and then the hinge pin to another mom and pop shop in that political district and then sending the two parts to another mom and pop structurial testing shop in another poltical district and then getting assembled in another .. on and on and on … was a selling point for the shuttle and one of the bragging points that the shuttle had the most parts etc….
this was exactly the wrong way to build affordable and sustatinable.
Especially when you are designing and building things based on prior systems, there are a lot of shortcuts through the procurement, test, verification regs. But when I look at the people at least from JSC who were put in charge of this job, I see that not a single one, Gerstenmaier, Geyer, Hanly, most of the underlings, had never designed or built anything before in their careers. Maybe that had something to do with their failure to get the job done on a reasonable schedule and for a reasonable cost?
You are misunderstanding the entire process. Congress wants this cost plus, fixed fee, sole sourced, FAR development nightmare to continue as it is now. It is doing EXACTLY what congress funded it to do.
Keep the lights on at all the NASA centers. Keep all the brooms being pushed, the computers churning out power points et cetera…
I have no doubt that Congress members all want their pork but that doesn’t preclude it from being designed and built on a reasonable time/schedule basis. Or are they purposely drawing it out from 6 years to 20? By this time and with these costs, as Space X has shown, there should have been an assembly line with vehicle being turned out every couple months. And, as we are seeing there is no shortage of launch vehicles to use prior to SLS.
If it was built on a reasonable time schedule they would need to already be funding what would have to be launched by it… There isn’t funding for that ..
You’ve made this point before, and while I generally agree that the desire to keep NASA operating (well, to keep the jobs, anyway), the choice of activities is an independent variable.
In other words, they coulda been actually building something efficiently. Then starting something else.
Nope. As I think more about this over the course of years looks to me like everyone in the process is so wed to ‘space is hard- and expensive’ that the costs are simply seen as the cost of business.
And in fact before SX I would have been at least partly sympathetic to the point.
Building something efficiently means you have to cut jobs, you have to streamline processes which means you have to cut jobs. The FAR contracting is a monster 50 years in the making at NASA and it is not going away soon. You saw the fights with people like Senator Shelby that kept hammering the point with commercial crew when NASA used SAA’s instead of FAR. It is congressional bread and butter the long term development programs.
Maybe people can learn the wrong lesson from mistakes. The Ariane 501 launch failure was, to a large extent, due to taking shortcuts with testing and verification. This was done because the software in question was copied from the earlier Ariane 4 launch vehicle, and had worked perfectly.
I think the lesson many people took away from this was never, ever, take the sort of shortcuts you mention. I think the correct lesson would have been to think it through and use good judgement about what shortcuts are appropriate and which are just dim. But “use good judgement” isn’t something you can write into an official practices and procedures document.
I still hear my parents using that same exact phrase overtime I screwed up as a kid: “you are not using good judgement”. And they were right.
Seriously? Isn’t this a matter of degree?
$16B and more than a decade to build a modern version of a well-understood technology?
Sure it’s complex. Yes, there are lots of little niggling things to fix. And in a world where SpaceX didn’t exist your argument might wash. But that’s a different world and the argument just doesn’t work.
Had your Cassini had even a portion of the delays and over-runs you’d still be dreaming about Saturn.
It’s definitely a matter of degree, and that was my point (although I may have said it badly.) It’s easy, especially when astronauts are involved, to get into a “failure is not an option” mindset. If an institution starts thinking that way, then people assume that, when a risk is identified, you should automatically spend money and time mitigating or eliminating it, regardless of the cost. I know people who would be appalled by the idea of simply saying, “we’ll just accept that risk; avoiding it would cost too much.”
I don’t agree with that. I think the optimal balance involves considering the costs and risks, and sometimes accepting risk. That’s especially true of unmanned spacecraft. Once astronauts are involved, the risk involves all sorts of intangible values. Those make cost-benefit analysis sort of nebulous.
SpaceX is, apparently, more willing to strike a balance than chase down every imaginable risk.
By the way, Cassini was, after 1992, under a fairly strict cost cap. A companion spacecraft (which would have been like the Rosetta mission, only a decade sooner and by NASA rather than ESA) was canceled due to cost overruns. Cassini dealt with this by descoping and dropping capabilities. But I definitely have seen, on Cassini and other JPL missions, an inclination to spend time and money on risk mitigation and an unwillingness to accept risk.
In one case, I heard a JPL engineer say he thought something was a potential risk, that he couldn’t prove it, but the subsystem in question was guilty until the relevant engineers could prove it innocent.
Risk wise, as initially designed, most of Orion is single string-no redundancies. They’ve lately been thinking this was dumb and they need to add some rendundancy beginning with the 2nd or 3rd mission. What will that do to costs and schedule? Did I miss it or was this addressed in the report?
re; JPL engineer: I take your point that this engineer was perhaps overly-cautious (that is your point?).
On the other hand, the ‘gut’ feeling of an experienced engineer? Priceless.
Stating the obvious it’s the inaccessibility of these systems after launch driving the caution. I wonder, if you allow, how these issues will be seen a century or more hence, in a world where a ‘broken Cassini’ isn’t such a big deal because HSF can send a repair man. Of course by then the bugs are surely well-understood, I suppose. Cassini is wonderful, even descoped, but in a century it will be seen as an automobile requiring a hand crank in a world of self-starting cars.
I do not believe that was the Orion, I think that was still the CEV?
As time has gone by, the CEV/Orion became smaller and less capable. Remember originally it was going to carry 7, have a deployable heatshield, land landing on airbags, be reusable. Orion represents the 3rd or 4th downsizing from the original plan. Also, as far as the US/Lockheed responsibility, its less now since ESA is providing the Service Module, which is well and good except the price and schedule did not improve.They lost at least half the vehicle and quadrupled the price.
Same thing happened when the US transitioned from Freedom to ISS. Bringing in more partners to share the costs only increases the integration costs and makes you dependent on another the schedule and funding coming from another country (or in the case of ISS, countries).
Rather than have them help deliver 70% of the mass: dirt cheap propellant allowing more risks to reduce costs with reuse.
It also reduces risk of LOM by decoupling propellant delivery flights from delivery of mission elements (i.e., elements stay on the ground until needed for mission)
Yes, bringing in more partners increased costs and delayed schedule. But then Freedom never generated any flight qualified hardware and in its last year it survived funding by only one vote. Need to re-package as ISS. Dang, fast forward to VSE which arrived but DOA as Vladislaw pointed out “…highly reliable and need only small ground crews.” Out with CEV and in with Orion which was only way to get funding.
I do not completely agree with this. From 1984 until 1993 Freedom included Canada, Japan and ESA. Only ones who were new to ISS in 1993 were the Russians. Russians had some effect on costs; they should not have all been cost increases. Soyuz meant that we did not need a CRV and for the last five uyears and the next 2, we don’t need a crew logistics carrier. Even at $75million a seat its a lot less than Shuttle. Russian docking module for Shuttle meant the US did not need to build one. US no longer needed a propulsion system. ISS went to a higher inclination and while the Russians instigated it, it was always dumb to have the station in a low inclination orbit when one of its prime tasks was going to be to fly over more of the Earth-particularly the countries paying for it. Russian delays with some modules increased costs a bit, but Russia saved the program-it probably would not have continued without them because Congress would have stopped the funding, and certainly after Columbia, Russians saved the program. US and Russian segments operate just about independently so I don’t see a lot of technical integration costs.
But at the time Canada, Japan, and ESA were not “on the critical path”. When Russia was added, it was placed square in the middle of the critical path. Russia provided the first “core” modules of ISS responsible for power, life support, attitude control, reboost, and etc. At the time there was much grumbling about how Russia wasn’t meeting schedules, which caused everyone else’s schedule to slip.
It was many years, and many shuttle assembly flights, later that the US side started to become more self sufficient.
And Russia is still the only provider of an emergency crew return vehicle for ISS. It will take a few more years before the US Commercial Crew program starts flying actual ISS crew rotation missions.
You are right, the Russians were on the critical path. Clinton was going to cancel ISS until he decided it had value for working with the Russians in the wake of the collapse of Communism. FGB was a little late-not too late since the US paid for it, did not have much effect on US costs. SM was later and did have an effect when permanent crew was delayed. Hard to understand how this effected costs since costs on ISS have been nearly flat at $3 billion/yr
The argument goes that because money was spent on engineering, communications, travel, and etc. in order to work with Russia (and integrate their modules and systems with US systems), that was taking away money which could have been spent on other things. In other words, it took money away from engineering US systems for the US modules, pushing their schedules further out into the future.
I’m not saying I completely buy that argument, but I hear it frequently. And it does fit nicely with the saying that “assigning more people on a late project makes it later”. I’ve experienced this first hand at my job. You end up spending more time trying to bring the new people on a project up to speed than you would have spent if you had just done all the work yourself. This is a mistake that middle management makes time and again because upper management demands that *something* be done to speed up the project.
Did not the CEV merely get named “Orion” as part of Constellation? Or did the name carry along (with the rest of Constellation) add-on riders that bulked it up in capability & cost? My memory is getting fuzzy.
Pre Griffin, the 60 day study and the ESAS, the guide was The Vision for Space Exploration. This said no new rockets for NASA. The CEV was going to go on an existing rocket from the EELV. After the “study” the EELVs were suddenly extremely dangerous, and to expensive and almost impossible to upgrade. It was never supposed to be more than a down and dirty capsule until commercial cargo and crew were finished.
If you read the VSE:
“In the days of the Apollo program, human exploration systems employed expendable, single-use vehicles requiring large ground crews and careful monitoring. For future, sustainable exploration programs, NASA requires cost-effective vehicles that may be reused, have systems that could be applied to more than one destination, and are highly reliable and need only small ground crews. NASA plans to invest in a number of new approaches to exploration, such as robotic networks, modular systems, pre-positioned propellants, advanced power and propulsion, and in-space assembly, that could enable these kinds of vehicles.”
So what would “these kinds of vehicles” look like? Space based, reusable, modules, assembled in space, flexible path, multi destination, that would use fuel depots.
Now does that sound like Orion?
To me that sounds like the Nautilus X.
Ahhhh, the timeline details are coming back. Thanks for the refresher.
Belatedly:
Vladislaw didn’t really address your question directly.
The name was merely the name. But “Orion” did become a different beast than CEV. CEV was intended to fly on EELVs. When Griffin insisted on Ares, he was putting a crew capsule on a modified SRB, that created safety issues that radically altered the design.
When a solid motor fails, it creates a cloud of solid pieces of burning fuel. Your abort escape system needs to get much, much further away from that debris than it does from a liquid fuel explosion. (Otherwise your parachutes will burn and you’ll die.) That dramatically increases the size of your launch abort motor. In the case of Orion, the LAS is nearly as heavy as the capsule itself. That immediately halves the mass available for the capsule-proper… which feeds back into the rest of the design.
Same thing happened with the MPCV, under Obama. Calling it Orion was just continuing the name from Constellation. But “Orion” was different from “MPCV”, for the same reasons that the old “Orion” was different from “CEV”.
MPCV was intended to be “Orion-lite”, a return to the idea of launching a small but flexible capsule to ISS (or another LEO-assembled vehicle) via EELV’s. But when Congress insisted on SLS, it caused the same issue as Ares I, for the same reason. So New Orion suffered the same inherent mass issues as Old Orion.
We’ll all revisit this when FH is flying and SLS is still a paper rocket.
I may be mistaken, but you’re confusing FH with the MCT. SLS has (I believe) ten times the Mars capacity than FH does. If that is correct, this comparison isn’t apt
I believe he is refering to the THROW WEIGHT of the Block 1 Space Launch System versus the Falcon Heavy to LEO. Since that is ALL the block 1 will be doing, putting hardware into LEO.
that is true. I guess in my mind, perhaps incorrectly, I think of the block II SLS when I talk about the program. As far as what has been released, there are no plans to have block I & II for FH. there are major differences between FH and the MCT
In effect, there are plans for a block II Falcon heavy, although it’s never discussed in those terms.
The Falcon Heavy/SLS (block I or II) comparison is very destination-dependent. Falcon looks pretty good in terms of mass to LEO. It doesn’t compare nearly as well for payload to Mars. (Although not an order of magnitude worse: Red Dragon would be about 7 tonnes, and an SLS isn’t going to land 70 tonnes.) The reason is that the Falcon doesn’t have a good, high-energy upper stage.
As of January, 2016, SpaceX has a Air Force contract to develop a engine for such an upper stage, and the Air Force seems interested in further development of the full upper stage. That would bring the Falcon payload to Mars much closer to that of SLS.
I would like to see some performance estimates, A methane-fueled Raptor-powered upper stage for the FH would be of particular benefit in BLEO trajectories, whether to GEO, the Moon, or Mars.
Work the problem backwards starting from space with no gravity. From the rocket equation, same dV and same payload fraction and delivered mass, the methane powered rocket full mass must be ~25% more than a LH2 full stage mass. IOW, the initial mass to LEO is 25% more (ISP of ~460 to 360s). The ideal *performance* architecture would launch LH2 from earth with methane/RP-1 rockets, and propellant, crew and supplies would be positioned using both chemical LH2 and EP tugs. The chemical transfer stage would be filled on orbit. With no gravity well missions, alternative fuels are minimized. This architecture mixes the best of two or more LVs.
I agree that for ISP nothing can beat LH2, but methane is a reasonable compromise between LEO and GTO requirements and apparently SpaceX believes it may minimize total cost by allowing a single architecture to be used for both types of payloads.
The total costs of the USG are the metric. Not the cost of DOD launches separate from NASA launches. If the LH2 engines were only 25% more than Merlins, then it would be right at the break even point.
The optimization depends on the average mT per year over a long time frame, the fixed and recurring costs, the cost of non-sole source.
Example 300mT of propellant to Mars launched to LEO by Falcons. 25% more for Methane transfer stage, 375mT. at 3000/kg thats 225M. 100mT would be 75M. So unless non sole source goes away, not sure why SpaceX believes methane/RP1 to Mars is the most cost effective solution.
I agreed, but Jupiter is also on my mind.
Hydrogen has better thrust to weight than kerosene. However, using launch velocity numbers when talking about flights to Mars might not be that relevent to two-way human flights. The only spacecraft that has ever went into solar orbit to a destination, orbited that destination, left and went back into solar orbit and then orbited something else had electric propulsion.
So if I put a $150 M something launch cost to LEO for a 50 meteric Ton electrical Mars transfer vehicle, and set it next to a $600 M something launch cost to LEO for a 70 or even a 130 metric ton electric Mars transfer vehicle…which one will get funded?
Actually, even with electric propulsion (which is a good point), I still don’t think mass to LEO is a good metric. Electric propulsion is inefficient and slow when it comes to going from LEO to Earth escape (or vice versa.) I think launch to a minimum escape trajectory (C3 of zero) might be a better comparison.
10x? Not even close…
http://www.thespacereview.c…
In payload size and throw weight, SLS…even the 2018 version…is significantly bigger than FH. But no where near 10x bigger.
Cost/launch for the FH vs SLS makes FH launches to Mars far more doable…if you want to factor that into their “Mars Capacity”. That’s up to you. Total cost to SpaceX for Red Dragon is projected to be $300M. NASA just signed an unfunded Space Act Agreement with them to do that one. It’ll cost NASA some $35M in internal costs and is intended to launch the same year as SLS first test flight.
$300M buys half an SLS launch to LEO. So in terms of $ per mile, with possibility being a cost issue…
That article only discusses block I. As posted later in this discussion, I had assumed block II
Block II is 8 something wide payload fairing and 130 metric tons to LEO.
https://en.m.wikipedia.org/…
FH is just under half of that at 54 metric tons. I realise there is more to Mars flight than just those numbers…still not 10x. That’s the beef that a lot of folks have with SLS…they want a reasonable cost to performance ratio and it just isn’t there. That is the point that I was trying to make.
Further, as the article points out, Block II isn’t a fair comparison since it is no less a paper rocket than the SpaceX mega-heavy lifters that they are drawing up. Once FH is in operation and Red Dragon flies, both the market for those megalifters, and the rockets themselves, will go into development on a faster time scale than NASA’s Journey to Mars.
In some countries, someone would be shot for this ridiculousness….smh…
A generation of engineers just flushed down the toilet.
Technically, the faith based part of SLS Orion planning is as much faith that budgets will arrive to continue to be sucked up into never ending SLS Orion work, due to cost plus inefficiency’s and all, as much as it is faith that still MORE budget will appear for using these systems, and making payloads for it all. Having used up all likely funds on the sports car, the “plan” is to pray money drops from the sky to take the road trip.
You have the sports car … the body.. but it is still sitting up on blocks waiting for the engine, the transmission and the sporty tires…
True.
Who takes any responsibility for this? So many years, it’s insane. Disgraceful really. I know the money is tight, but NASA seems only a shadow of what it once was. Are there no politicians to even raise a ruckus about the years this is taking. Does no one in Houston or Washington even really care?
The people Congress do care – about federal budget dollars getting funneled into their districts.
NASA does the best it can with what it gets.
My observation of Orion over the last dozen years, and before that of ISS, is that the program offices have become far too powerful and immature inexperienced program managers make really dumb calls on requirements, process, design. You also do not have the NASA expertise from across the agency in your camp helping to make the proper technical calls.
I should qualify this by saying that until the 1990s, there were experienced NASA technical experts. Since then NASA has aged (not matured) and there is little expertise left because the programs have been in control for too long. This may have something to do with Goldin kicking out all the Freedom expertise when the Russians came into ISS. That was a dumb and shortsighted move. Now of course we have all the neophytes coming in and there is no one to train them. Maybe they should just be put in charge because the managers have sure screwed it up good.
I saw they had a training class the other day out of safety “what to do when you see the program manager making stupid decisions”.
I know my attitude has become “you’re the program manager, you have the budget, you have the authority and responsibility, they put you in charge so I guess they (someone) must have some confidence in your ability (I don’t see it myself) so if you want to make stupid (costly) decisions, go ahead.”
The technical organizations should have the technical experts that oversee the technical work. The program office ought to managing requirements, cost and schedule, not the technical work.
–
I suspect that the final lesson of SLS and Orion will be this: Vertical integration is not only more efficient, it is the only way to get a good intersection of smaller budget and rapid completion. I suspect that, should SLS have a replacement in the ‘government rocket’ stable, the RFP will specify as vertical an integration strategy as the bidder can achieve.
Vertical integration by a government owned and operated launch vehicle is quite simply never going to happen. Politics necessitates that money spent on large government programs is spread over as many Congressional districts as possible. This means spreading sub-contracts around to as many companies, located in as many states, as possible.
The only way to get vertical integration in a launch vehicle used by the government is to buy launch services from the lowest bidder.
Put another way, NACA contributed greatly to early aviation in the US, but it never owned its own airline. So, why should NASA own its own launch vehicle? Why can’t NASA act more like its predecessor NACA and stick to developing new technologies for use by US launch providers?
Isn’t that exactly where it is headed, for the most part? NASA has pretty much abandoned LEO in favor of SpaceX et. al. SLS, sure, is an in-house rocket, but that’s because at the time it was initiated there was no real way to get large payloads BLEO.
Things change with time. And while I don’t think that SLS will fly at all, even if it does it will be abandoned in the next decade or so as FH and FH+ become reliable and cheap.
So, you heard it here first: by, say, 2030, NASA will be out of the in-house rocket biz.
Or not.