Anyone who has been watching Isaacman’s interviews or reading his tweets knows SLS/Orion days are numbered. Dwelling on the past isn’t important, with the exception of not making the same mistakes.
Starship will will be busy but will have a lot of excess capacity for customers to use. How NASA and business chooses to use it is yet to be determined.
Dwelling on the past is important for understanding the present and how we got here, which is what the article does so well, perhaps though as “this is why we’re still stuck here, and we need to quit repeating the same failures.”
Right now what I’m not seeing from NASA, though it may be coming soon due to Isaacman, is a shift towards exploring what they can do with the new opportunities created by Starship and Superheavy. Hopefully lots of big brains at NASA are spending time (off the books, of course) madly doodling new configurations, special modifications, and all kinds of fancy things they want to put in Starship’s cargo bay, or launch on top of Superbooster.
Before, there wasn’t all that much impetus for a lot of experimental propulsion projects because there was no way to afford launching them to LEO for a test, absent shouldering out a bunch of other potential projects. But if Musk succeeds in lowering launch costs by orders of magnitude, all kinds of new, cutting edge missions because viable. And by viable, I mean going beyond a paper study and actually ending up in space.
And there’s so much design space that simply opens up when a new rocket design is rolling out to the test stand all the time, as opposed to flying the same exact vehicle for decades. Heck, a decade from now the retrospective view might be that the most critical advance was building vehicles out of welded stainless steel, allowing modifications to be made unimaginably faster and simpler, and thus empowering the rapid fly, re-design, fly iterations.
.
Heck, one simple thing I’ve thought of regarding a multi-stage re-usable launcher is to simply split the stack with diagonal cuts instead of horizontal cuts between the stages, so that front and rear ends of a stage are sloped instead of flush, making it much easier to shield the ends through a pancake (skydiver) re-entry. Such a simple thing, yet I’ve never even seen it hinted at anywhere.
There seem to be at least some in the NASA Science Mission Directorate already working on Starship-enabled exploration missions. May their tribe increase.
I must say the idea of a launch vehicle sliced like a so-called “Italian-cut” green bean is a novelty. Starship achieves the desired effect, absent such a bias cut, by enclosing all of its engines within a skirt that extends slightly past its engine nozzles.
The short skirt protects the engines, but the idea occurred to me when I was thinking of having Starship launch a third stage, probably cryogenic, for a longer reach and a better fuel fraction getting to LEO.
That third stage is easy. The problem is how to keep the nose of the second stage from burning up on re-entry, since Starship uses a large positive angle of attack (nearly but not quite flatplate) so the re-entry heating never directly impinges on the aft engine bay. Starship’s nose, of course, can’t use that trick and is covered with tiles. And to make sure plasma doesn’t enter the engine bay, the re-entry is flown at usually a 70 degree angle-of-attack, though sometimes it noses over to 60 degrees.
So that means they want a 20 degree angle between the angle-of-attack and the face of the skirt and the complex piping and wiring inside the bay. So if Starship was just a second stage, the front is going to look much like the top of Super Booster. Following the design of the engine bay aft, the flat face of the forward section needs to slope back 20 degrees from the airstream, and since sometimes the angle of attack is lowered an extra 10 degrees, that slope would need to be 30 degrees.
And that would mean the back end of the third stage is sloped at 30 degrees, so if it re-entered, an angle of attack of 90 degrees would be fine for it’s aft engine bay.
And following the pattern, a stack with a bunch of sections would have each stage’s ends cut at a 30 degree angle, front and back, but the angles would be opposite instead of parallel, so the each stage would be oriented with it’s heat shield on the opposite side of the adjoining stages. Only the topmost stage would need a conventional nose, and that’s for boring up through the atmosphere.
And thus you could get a fully re-usable three stage rocket, or one where you re-use the first two stages while the third is a deep space vehicle that doesn’t return.
Hrm… That opens up another possibility for the second stage to ditch the forward flaperons and have a single flap that hinges forward and out from the sloping nose/interstage section after separation, to vary the amount of drag on the forward end for pitch control. With two of them, one could also do some roll and yaw control like they do now with the forward flaperons. Basically, the sloped interstage area would be a convenient place to stow them during launch, getting rid of the need for the complex thermal issues with the exiting exposed hinge location, and reducing drag on the way up.
Both interesting ideas that some enterprising soul should put to a test in a wind tunnel somewhere. The idea, if I understand it properly, is to have the 2nd stage, with the bias-cut nose, re-enter with its long side down and its short side and elliptical “cut” up.
Absent any forward control surfaces, I worry that the lowest-energy roll orientation could well be the opposite of that intended – long side up and short side and ellipse down. Not good. Your notion of swing-out control surfaces probably ought to be regarded as a mandatory feature of this configuration to prevent that.
It also seems as though at least the forwardmost part of the ellipse surface should be TPSed in some fashion for the same reason that the nose of Starship is tiled all the way around at its very top.
Good catch. Absent control surfaces, it would want to roll upside-down because the long (bottom) side will have more drag than the short (top) side. So it would either need to deploy flaperons, or perhaps the equivalent of vertical stabilizers on both ends, somewhat akin to how Spaceship One rotates its wings so they become like shuttlecock feathers to keep it oriented bottom down during re-entry. Simple flaperons fore and aft, since they’d likely hinge near the bottom, would just add to the roll instability. Lowering the CG (moving it towards the long side) might help, but the only moveable internal mass would be the land fuel, which would normally just roll around in the fuel tank unless there was a special header tank running along the long side.
It’s solvable, but so far I don’t think the solutions would be elegant, with big tile-coated surfaces sprouting from both ends.
And I agree about the forward lip getting a lot of TPS, similar to how a regular capsule gets special attention where the lower frustum meets the bottom shell.
Once starship is flying, NASA and others will be better able to plan for it. Too bad NR-1 wont be flying on a starship but that will be something new if they are able to do it.
Very creative idea. You should get yourself a 3d printer
Hot staging would require a side kick to counteract the significant torque on both stages due to the angled intersection between them.
Anyway, I’m wondering about having a hoop on each end, basically a circumferential ring of high temperature alloy covered with tiles, that pivots out from the long side at what will be the bottom during re-entry. They might stability the roll instability, but they’d need an additional degree of freedom for yaw. Hinging in an out, required to extend them, would already cover pitch.
I’m thinking of a ring instead of arms or flaps or insect antenna because a ring wouldn’t get in the way of engines or anything, and would naturally be out of the way during hot staging.
Mr. Zapata’s capsule history – with much of it being the history of one particular capsule – was a good effort. Where his piece falls short is in its vision of the future. It’s still a NASA-centric one. Not unexpected from a career NASAn, but also wildly at variance with the future the foundations of which are already being poured – literally – at Starbase, KSC and Canaveral.
There remains a notable tendency on the part of old hands from both NASA and the legacy contractor milieu to continue to ignore, almost entirely, what is going on outside of their familiar bubble. Plains Indians of the early 16th Century could be pardoned for regarding travelers’ tales of pale-skinned strangers with metal armor and firearms appearing in far-off Florida as fanciful fictions. Contemporary NASA and legacy contractor personnel have much less excuse for entertaining such dismissive attitudes, especially Space Coastians of traditionalist employment as they are cheek-by-jowl with NewSpace on a daily basis. Yet many, even in Florida, appear to persist in a “nothing to see here, move along” mindset.
Particularly jarring – and pathetic – was Mr. Zapata’s sneer at “zeros and ones” in the same sentence in which he had already defined Starlink as a product. Starlink is an infrastructure. The product that moves on it is entirely those zeros and ones. The same will be true of Starlink Mobile, SpaceX’s just-announced effort at becoming a global telco behemoth in its own right in addition to its current status as a global Internet service provider behemoth. And then there are the SpaceX Earth-orbital and Moon-centric AI data center initiatives – additional infrastructures, the products of which will also be entirely zeros and ones.
Those zeros and ones are what will pay our way to the stars. The revenue from any physical products made in space and destined for Earthside markets will, comparatively, look like the contents of a tip jar.
Mr. Zapata does a good job of limning the potential difficulties of sunsetting SLS and Orion in their entireties, though he at least acknowledges that Administrator Isaacman has made a very decent start on that job via canning EUS, ML-2 and Gateway. For those achievements alone, Isaacman’s political acumen has now been demonstrated to meet or exceed that of any previous holder of his office. And I think he’s just getting warmed up.
Despite his quickly-acquired – and well-deserved – reputation as a teller of uncomfortable truths, Isaacman also obviously understands the virtues of strategic ambiguity. Gateway is not cancelled, – perish forbid – just “paused.” To paraphrase an early Saturday Night Live catch-phrase, “Generalissimo Gateway is still paused.”
When one’s heart has already been marked as a donor organ for another project – as Gateway’s has in the form of its Power and Propulsion Element – you are not “paused,” you are dead.
As to what remains of SLS and Orion, Isaacman has made it clear that he’s going to allow their contractors every opportunity to advance their cadence of march to double-time without being needlessly specific, at this point, about what the consequences will be for failure to proceed with sufficient dispatch. Rather than bald threats now, Isaacman reserves the ability to act more in sorrow than in anger later when SLS and Orion all but inevitably fall short.
Or, we could all be surprised and find that the SLS and Orion “horses” can actually be taught to sing.
As to the mechanics of the still-under-revision Artemis Moon mission calendar, there has been some talk of not using the last ICPS upper stage on the Artemis 3 mission so as to save it for later and allow more time for the Centaur-V-for-ICPS swap-out to be engineered and accomplished. If some suitable placeholder spacer can be ginned up to stand-in for ICPS by next year, that certainly looks doable from a vehicle performance standpoint.
I am of the opinion that SpaceX will demonstrate large-scale on-orbit propellant transfer from a tanker Starship to a LEO-orbiting propellant depot Starship before the end of this year. It is even likely that two or more such tanker missions will be run to the same depot ship in 2026.
I am also of the opinion that the initial version of SpaceX’s HLS Starship will be ready to go in time for the Artemis 3 mission at its now-planned mid-2027 target date. Blue Origin’s initial manned lunar lander, I fear, will be a no-show – or perhaps a partially-complete show analogous to the incomplete Orions flown on Artemis 1 and 2. A complete Blue Origin lander of some Mk 1.5 configuration in time for Artemis 3 would be a surprise to me, but a welcome one.
To provide the best possible chance for an early-2028 Artemis 4 initial landing mission to proceed as currently scheduled, it is necessary that SpaceX do its unmanned full-refill, LEO departure, transit, landing, surface loiter and ascent test of HLS Starship as soon as possible. The way to do this, I think, is to keep doing tanker missions to the LEO depot ship until, and even beyond, Artemis 3, as necessary. Then, when an adequately full depot is achieved, refill the HLS Starship lander left in orbit after Artemis 3 and send it Moonward as the lunar landing test article.
Loren Grush recently had a story on Bloomberg News that alleged Artemis 4 might also excuse ICPS from the chore of trans-lunar injection, with the SpaceX HLS Starship docking to Orion in LEO and then proceeding Moonward as a joined unit. Seemingly little-noted was the fact that, if ICPS is relieved of its main function, it might just as easily be relieved of any necessity to show up at all. Thus, we could have two consecutive SLS missions – Artemises 3 and 4, respectively – absent any SLS second stage. This would allow the last ICPS to be used for Artemis V if Blue Origin’s lunar lander is ready – and SLS as well, of course – by late 2028. I do not anticipate that said lander would be able to do the TLI burn in place of ICPS as HLS Starship can.
Under this scenario, we could get all the way through Artemis 5 without needing to fly an SLS with a Centaur V-based upper stage.
As of now there are no more SLSes and Orions mandated for missions beyond Artemis 5. If Artemis 5 slips from its current aspirational launch date of “late 2028,” I can certainly see Isaacman officially playing Taps over both programs before leaving office. That would also mean cancelling the “standardized” SLS with the Centaur V-based upper stage without it ever having flown, but we already have a precedent for that in the form of EUS.
This all assumes – as I very much do – that, by this time, SpaceX will have debuted and tested a crew-capable version of Starship that can take-off from Earth, refill in LEO, then both transit to and from the near-lunar vicinity including a direct re-entry to Earth.
Blue Origin, I expect to also be working on something able to take people to and from the Moon in place of SLS and Orion. Perhaps that will, at least, initially, be Orion riding some version of New Glenn. Isaacman, I’m sure, would be amenable to a stay-of-execution for Orion under such circumstances. That would tide Blue over until it can come up with something better, and far more reusable, that can also fly on New Glenn – probably the 9×4 variant.
Heck, stretching out the executions in this fashion most likely increases the probability that all can be accomplished with the available store of political capital in-hand.
I think Orion/SLS is going to be an exercise in trying to find more delta V somewhere that can be shoehorned into making a lunar mission workable, since the math doesn’t work out with the existing hardware, and attempted architectural fixes like Gateway are massive detours.
As for the broader view of NASA, there are several fundamental changes that hit space flight in quick succession, and I think some of the more important ones stem from SpaceX becoming a commercial success beyond anyone’s imagining. Now there is a launch provider with its own huge and independent revenue stream. Deeply entwined with that success is that it’s very nimble and bold. In competitive terms, it has a very fast OODA loop.
Isaacman has dramatically improved the speed of NASA’s decision making, but he’s got a long way to go and slimming a giant government bureaucracy will not be easy or quick. Even when NASA tries to be bold, that boldness is going to have long timelines because it had to come from long studies, proposals, budget requests, bids, contracts, milestones, design reviews, etc. Basically lots and lots and lots of meetings with lots of officials having to sign off on each step. Meanwhile Elon’s OODA loop is sometimes him just saying “Yeah, lets try that,” and a few months later there’s something new rolling out. And the startups, including many Chinese ones, are trying to follow SpaceX’s approach because smart entrepreneurs emulate success instead of stagnation.
So NASA will likely continue to lag and have some problems making workable plans that interface with the faster, nimbler companies unless they keep old designs around as “legacy hardware” to fulfill NASA requirements from five or six budget cycles earlier. Many of the NASA initiatives will probably have been leapfrogged or bypassed before they are ready to implement.
If that starts proving to be the case, what Isaacman or his successors might have to do is switch to a different, looser contracting model where it defines national goals and priorities that need to be addressed, working with the private sector to see that someone executes the missions. Perhaps a future debate will be why NASA needs to be so big and have so many employees if its role has devolved into setting safety standards and handing out X-Prizes.
You’ve put your finger directly on the crux of the problem in deciding what NASA should be and do, going forward.
Pretty obviously, leadership in the engineering of launch vehicles has already long-since passed to the private sector and the same is in the process of happening to both unmanned and manned spacecraft destined for the Moon and Mars.
I think manned spacecraft for any destination or purpose will be strictly private sector-provided from now on as will an increasing percentage of things like unmanned landers. There will be a lot more of the latter carrying freight than science instrument packages.
As the vast majority of people working in space will have company logos on their coveralls instead of NASA ones within a decade or two, astronaut training will change greatly to cover all of the different types of workers employed extraterrestrially and will be handled by employers, not NASA.
There may remain a few niches for traditional test pilot Right Stuff types, but not many. With future generations of both military and civilian aircraft destined to be unmanned, test pilots, like other types of pilots, will join horse cavalrymen and armored knights in the ranks of formerly essential and valorous vocations that have been rendered obsolete by the relentless march of technology.
NASA will, in pretty short order, be left with only the engineering of pure science missions to the outer planets, the Kuiper Belt and – eventually – the Oort Cloud. And even these projects will increasingly rely on technology developed by the private sector.
NASA seems inevitably destined to shrink quite a bit. Operating test facilities like Stennis and Plum Brook or the wind tunnels at Ames might be viable, longer-term, but even that is far from assured. SpaceX, for example, has its own test facilities at McGregor, TX and has never used Stennis. A number of the newer NewSpace launcher firms still use Stennis, but not even all of them are on-board there.
Plum Brook could also find itself obsoleted. As launch costs diminish, it may soon be cheaper just to put prototype hardware into actual space than to expensively transport it to Ohio and use simulation chambers that are expensive to rent because they are expensive to operate and maintain.
At some point – perhaps as soon as mid-century – it may make more sense to just shutter NASA entirely and transfer whatever science efforts it still pursues to the National Science Foundation.
I agree. I could easily see NASA hewing more to the roll that it had as NACA, and I can see it maintaining a vital role in space-based astronomy and planetary probes (JPL), and perhaps as a major body doing lots of the research on human physiology in space, space medicine, and perhaps as a go-to regulatory authority to try to keep LEO from becoming total chaos.
Basically, cover the needs of government and industry on standards, regulation, and fundamental research, and whatever aspects of science that won’t attract private capital. Formerly, that was pretty much most of space vehicle and space flight research (along with the military) because there was no significant private market until satellite communications and perhaps weather satellites opened one.
One change we might already be seeing is the massive increase in the number of young engineers who are gaining expertise in satellite and space vehicle design and operation. Starlink alone has 3000 employees, and they’ve worked tirelessly to not only design some incredible satellites and constantly improve and rethink them, they’ve also gone beyond that and figured out how to mass produce them commercially, moving well beyond things like assembling bespoke spy satellites hand made by Trappist monks, each with three PhDs. The same is happening at Blue Origin, though in a deep underground bunker somewhere.
And those engineers are also learning or mastering space capsules, space suits, life support systems, control systems, and all the rest. They’ve mastered their own mission control systems, launch systems, and recovery systems.
We’re going to have a space workforce packed with a whole lot of highly capable and experienced young engineers, and just the existence of them should start fertilizing new start ups, similar to what happened in Silicon Valley as all the cool young nerds learned digital electronics and computing. It created a new ecosystem. Perhaps there are parallels to the shift from the mainframe era of IBM, Burroughs, Sperry, Univac, NCR, and Honeywell to what followed, as computing went from a boutique speciality to a ubiquitous everything.
Generally agree, but NASA should not be given any sort of regulatory authority. Particularly not for things like LEO traffic management which is already pretty much a solved problem. It’s been solved by SpaceX and said solution has been offered for free to other satellite constellation operators on a co-operative basis.
The only entities not likely to get onboard are Russia and the PRC. As I don’t see either regime – or either nation, for that matter – lasting much longer, this will be strictly a short-term irritant anent rational space traffic management – especially after every Earth orbit is swept clean of their legacy junk once they’re both gone.
SpaceX has already forgotten more about space traffic management than NASA is ever going to know. And, as there is not a damned thing NASA could do about Russia and the PRC declining to play nicely with others in orbit, there is absolutely no reason to charge it with regulating the spaceways. NASA simply isn’t up to the job.
SpaceX having a glory-days-of-Silicon-Valley affect on the aerospace industry generally is already very much a thing. SpaceX alumni can be found everywhere in more recently founded NewSpace companies – many in founder and senior management roles. The size of the “SpaceX Diaspora” powering all of this ferment will only increase as the company’s workforce continues to expand.
DE writes:
” Where his piece falls short is in its vision of the future. It’s still a NASA-centric one.”
NASA has not made the mental leap to moving from service provider to service consumer.
And even Isaacman encourages being stuck in the past when he says he wants to re-acquire that Good Old Expertise.
I’m not sure if Isaacman really believes that or if he is just saying that to mollify the masses as he puts his changes in place. I’m hoping for the latter and I do see some indication that he is going to make the switch. e.g. Gateway Pause.
That crew and capsule recovery evolution was embarrassing. Only because I witnessed it can I imagine ways to have made it more complex. But if I had to design it, I don’t think I could have found as many ways to waste time and money as NASA accomplished in the 2+ hours it took to just extract the crew.
One offshore installation vessel with a 100 ton crane, and they would have had the capsule and crew on a dry deck in 30 min, and then aboard 1 helicopter for the flight back to land within those 2 hours.
Yeah, that was embarrassing to watch. Not a good-looking way to finish the show. If Artemis 2 can be compared to a vaudeville act, that Keystone Kops recovery display would be the rough equivalent of the Irish Tenor’s pants falling down as he’s taking his bow.
Your description of the reasonable alternative is spot-on and could be shortened to just – “do it the way SpaceX does it.”
Obviously NASA didn’t rethink things during the years since Apollo, and just dusted off the old 1960’s documentation on how to recover a capsule after splashdown.
So perhaps it’s surprising that NASA used the Murtha, LPD-26 an amphibious transport dock, instead of the USS Hornet, CVS-10, which recovered Apollo 11 and 12, which they left sitting idle in Alameda California. The Intrepid and Yorktown are on the east coast, and could be made available for Atlantic Artemis splashdowns. Real space missions use Essex class carriers.
As I watched the recovery once the helicopters were on board the ship, I noticed that NASA only had one guy who had to escort each astronaut from a helicopter to a hatch, and then he’d have to go back to the helicopter to get another astronaut, and another. Obviously NASA needs to employ three additional escorts to speed things along. And why wasn’t anyone playing trumpets or throwing confetti? The recovery was badly understaffed, which wouldn’t be the case if they’d use an Essex class carrier with 2600 men, instead of just 550 people on an LPD. So what if SpaceX’s recovery ship only uses forty people?
Anyway, it’s odd that it takes so many experts and medical specialists to get astronauts out of a capsule during a normal recovery, yet if the capsule makes an emergency splashdown the astronauts are good to climb out of the capsule on their own, swim to an island, and live by hunting marine reptiles with a spear for a couple of weeks until they’re located. I’m sure that if left on their own, they’d eventually motivate and start doing things themselves.
Heck, now that Starlink is moving into providing global phone service, maybe we could just have the capsule come down where ever, and then the astronauts go online, call for an Uber, and make arrangements with a shipping company to freight the capsule back to Florida, so that the recovery force doesn’t actually employ anybody.
Given that three of the crew were F-18 pilots, it occurs to me that F-18 pilots routinely pull more G’s than the Orion’s max of 3.9 G’s, eyeballs in, and F-18 pilots go through far more violent maneuvers, and almost random ones at that. High G’s, shaking, bouncing, getting slammed around in the cockpit, negative G’s, high spin rates, and varying cockpit air pressures. Heck, during the arrested landing, an F-18 pilot goes through more G’s eyeball-out than an Artemis crew feels eyeballs-in.
So what if the Navy did F-18 recoveries like NASA does Artemis recoveries? First they’d let the pilot sit in the cockpit for half an hour or so, then send in a medical specialist to check him out. Then a team of medical specialists would lift him out of the cockpit and gently lower him to the deck, and then sit with him for half an hour before escorting him to sick bay for further tests.
But after a century of naval aviation, this would strike everyone as absurd. We know the fighter pilots are fine after recovery, even if their teeth might have gotten a little rattled.
With the first astronauts it made sense to have doctors rush in to get the freshest possible blood samples to see if weightlessness had any weird effects on them, and run them through batteries of tests to look for changes induced by their longer and longer stays in zero-G. But we’ve now recovered over 700 people from space missions, and not a single one has just keeled over dead after landing. They’ve all come through fine. We don’t need to keep treating them like the Mercury Seven or Gemini crews. We don’t need to keep treating them like lab rats who might hold the key to understanding space flight, and we don’t need to give each one a ticker tape parade through Times Square, either. Just because we started out doing it doesn’t mean we have to keep doing it.
Agree again, though the percentage of space travelers with air combat maneuvering and carrier take-offs and landings on their resumes will dwindle sharply in coming years but that will be compensated for by gentler rides both up and down.
A Starship will subject its passengers to no more G-forces on ascent than those experienced by today’s Orion, Starliner and Dragon riders. The ride uphill should also be notably smoother as Starship has no giant flaming organ pipes attached to rattle passenger teeth and it has way more engines on each stage than a Falcon 9. On both stages, said engines are also much farther away from the passenger accommodations.
On return, Starship subjects its passengers to no more G-force than Shuttle. The only thing even faintly alarming might be the flip-up maneuver prior to chopstick catchment and this will be mild even compared to typical amusement park rides.
Anent ticker tape parades, those are no longer possible due to both the modern equivalent of stock tickers – Bloomberg terminals – producing no paper tapes and to the fact that windows in modern high-rise buildings don’t open to allow any notional ticker tape or other forms of paper to be thrown.
Anyone who has been watching Isaacman’s interviews or reading his tweets knows SLS/Orion days are numbered. Dwelling on the past isn’t important, with the exception of not making the same mistakes.
Starship will will be busy but will have a lot of excess capacity for customers to use. How NASA and business chooses to use it is yet to be determined.
Dwelling on the past is important for understanding the present and how we got here, which is what the article does so well, perhaps though as “this is why we’re still stuck here, and we need to quit repeating the same failures.”
Right now what I’m not seeing from NASA, though it may be coming soon due to Isaacman, is a shift towards exploring what they can do with the new opportunities created by Starship and Superheavy. Hopefully lots of big brains at NASA are spending time (off the books, of course) madly doodling new configurations, special modifications, and all kinds of fancy things they want to put in Starship’s cargo bay, or launch on top of Superbooster.
Before, there wasn’t all that much impetus for a lot of experimental propulsion projects because there was no way to afford launching them to LEO for a test, absent shouldering out a bunch of other potential projects. But if Musk succeeds in lowering launch costs by orders of magnitude, all kinds of new, cutting edge missions because viable. And by viable, I mean going beyond a paper study and actually ending up in space.
And there’s so much design space that simply opens up when a new rocket design is rolling out to the test stand all the time, as opposed to flying the same exact vehicle for decades. Heck, a decade from now the retrospective view might be that the most critical advance was building vehicles out of welded stainless steel, allowing modifications to be made unimaginably faster and simpler, and thus empowering the rapid fly, re-design, fly iterations.
.
Heck, one simple thing I’ve thought of regarding a multi-stage re-usable launcher is to simply split the stack with diagonal cuts instead of horizontal cuts between the stages, so that front and rear ends of a stage are sloped instead of flush, making it much easier to shield the ends through a pancake (skydiver) re-entry. Such a simple thing, yet I’ve never even seen it hinted at anywhere.
There seem to be at least some in the NASA Science Mission Directorate already working on Starship-enabled exploration missions. May their tribe increase.
I must say the idea of a launch vehicle sliced like a so-called “Italian-cut” green bean is a novelty. Starship achieves the desired effect, absent such a bias cut, by enclosing all of its engines within a skirt that extends slightly past its engine nozzles.
The short skirt protects the engines, but the idea occurred to me when I was thinking of having Starship launch a third stage, probably cryogenic, for a longer reach and a better fuel fraction getting to LEO.
That third stage is easy. The problem is how to keep the nose of the second stage from burning up on re-entry, since Starship uses a large positive angle of attack (nearly but not quite flatplate) so the re-entry heating never directly impinges on the aft engine bay. Starship’s nose, of course, can’t use that trick and is covered with tiles. And to make sure plasma doesn’t enter the engine bay, the re-entry is flown at usually a 70 degree angle-of-attack, though sometimes it noses over to 60 degrees.
So that means they want a 20 degree angle between the angle-of-attack and the face of the skirt and the complex piping and wiring inside the bay. So if Starship was just a second stage, the front is going to look much like the top of Super Booster. Following the design of the engine bay aft, the flat face of the forward section needs to slope back 20 degrees from the airstream, and since sometimes the angle of attack is lowered an extra 10 degrees, that slope would need to be 30 degrees.
And that would mean the back end of the third stage is sloped at 30 degrees, so if it re-entered, an angle of attack of 90 degrees would be fine for it’s aft engine bay.
And following the pattern, a stack with a bunch of sections would have each stage’s ends cut at a 30 degree angle, front and back, but the angles would be opposite instead of parallel, so the each stage would be oriented with it’s heat shield on the opposite side of the adjoining stages. Only the topmost stage would need a conventional nose, and that’s for boring up through the atmosphere.
And thus you could get a fully re-usable three stage rocket, or one where you re-use the first two stages while the third is a deep space vehicle that doesn’t return.
Hrm… That opens up another possibility for the second stage to ditch the forward flaperons and have a single flap that hinges forward and out from the sloping nose/interstage section after separation, to vary the amount of drag on the forward end for pitch control. With two of them, one could also do some roll and yaw control like they do now with the forward flaperons. Basically, the sloped interstage area would be a convenient place to stow them during launch, getting rid of the need for the complex thermal issues with the exiting exposed hinge location, and reducing drag on the way up.
Both interesting ideas that some enterprising soul should put to a test in a wind tunnel somewhere. The idea, if I understand it properly, is to have the 2nd stage, with the bias-cut nose, re-enter with its long side down and its short side and elliptical “cut” up.
Absent any forward control surfaces, I worry that the lowest-energy roll orientation could well be the opposite of that intended – long side up and short side and ellipse down. Not good. Your notion of swing-out control surfaces probably ought to be regarded as a mandatory feature of this configuration to prevent that.
It also seems as though at least the forwardmost part of the ellipse surface should be TPSed in some fashion for the same reason that the nose of Starship is tiled all the way around at its very top.
Good catch. Absent control surfaces, it would want to roll upside-down because the long (bottom) side will have more drag than the short (top) side. So it would either need to deploy flaperons, or perhaps the equivalent of vertical stabilizers on both ends, somewhat akin to how Spaceship One rotates its wings so they become like shuttlecock feathers to keep it oriented bottom down during re-entry. Simple flaperons fore and aft, since they’d likely hinge near the bottom, would just add to the roll instability. Lowering the CG (moving it towards the long side) might help, but the only moveable internal mass would be the land fuel, which would normally just roll around in the fuel tank unless there was a special header tank running along the long side.
It’s solvable, but so far I don’t think the solutions would be elegant, with big tile-coated surfaces sprouting from both ends.
And I agree about the forward lip getting a lot of TPS, similar to how a regular capsule gets special attention where the lower frustum meets the bottom shell.
Once starship is flying, NASA and others will be better able to plan for it. Too bad NR-1 wont be flying on a starship but that will be something new if they are able to do it.
Very creative idea. You should get yourself a 3d printer
Hot staging would require a side kick to counteract the significant torque on both stages due to the angled intersection between them.
Anyway, I’m wondering about having a hoop on each end, basically a circumferential ring of high temperature alloy covered with tiles, that pivots out from the long side at what will be the bottom during re-entry. They might stability the roll instability, but they’d need an additional degree of freedom for yaw. Hinging in an out, required to extend them, would already cover pitch.
I’m thinking of a ring instead of arms or flaps or insect antenna because a ring wouldn’t get in the way of engines or anything, and would naturally be out of the way during hot staging.
Mr. Zapata’s capsule history – with much of it being the history of one particular capsule – was a good effort. Where his piece falls short is in its vision of the future. It’s still a NASA-centric one. Not unexpected from a career NASAn, but also wildly at variance with the future the foundations of which are already being poured – literally – at Starbase, KSC and Canaveral.
There remains a notable tendency on the part of old hands from both NASA and the legacy contractor milieu to continue to ignore, almost entirely, what is going on outside of their familiar bubble. Plains Indians of the early 16th Century could be pardoned for regarding travelers’ tales of pale-skinned strangers with metal armor and firearms appearing in far-off Florida as fanciful fictions. Contemporary NASA and legacy contractor personnel have much less excuse for entertaining such dismissive attitudes, especially Space Coastians of traditionalist employment as they are cheek-by-jowl with NewSpace on a daily basis. Yet many, even in Florida, appear to persist in a “nothing to see here, move along” mindset.
Particularly jarring – and pathetic – was Mr. Zapata’s sneer at “zeros and ones” in the same sentence in which he had already defined Starlink as a product. Starlink is an infrastructure. The product that moves on it is entirely those zeros and ones. The same will be true of Starlink Mobile, SpaceX’s just-announced effort at becoming a global telco behemoth in its own right in addition to its current status as a global Internet service provider behemoth. And then there are the SpaceX Earth-orbital and Moon-centric AI data center initiatives – additional infrastructures, the products of which will also be entirely zeros and ones.
Those zeros and ones are what will pay our way to the stars. The revenue from any physical products made in space and destined for Earthside markets will, comparatively, look like the contents of a tip jar.
Mr. Zapata does a good job of limning the potential difficulties of sunsetting SLS and Orion in their entireties, though he at least acknowledges that Administrator Isaacman has made a very decent start on that job via canning EUS, ML-2 and Gateway. For those achievements alone, Isaacman’s political acumen has now been demonstrated to meet or exceed that of any previous holder of his office. And I think he’s just getting warmed up.
Despite his quickly-acquired – and well-deserved – reputation as a teller of uncomfortable truths, Isaacman also obviously understands the virtues of strategic ambiguity. Gateway is not cancelled, – perish forbid – just “paused.” To paraphrase an early Saturday Night Live catch-phrase, “Generalissimo Gateway is still paused.”
When one’s heart has already been marked as a donor organ for another project – as Gateway’s has in the form of its Power and Propulsion Element – you are not “paused,” you are dead.
As to what remains of SLS and Orion, Isaacman has made it clear that he’s going to allow their contractors every opportunity to advance their cadence of march to double-time without being needlessly specific, at this point, about what the consequences will be for failure to proceed with sufficient dispatch. Rather than bald threats now, Isaacman reserves the ability to act more in sorrow than in anger later when SLS and Orion all but inevitably fall short.
Or, we could all be surprised and find that the SLS and Orion “horses” can actually be taught to sing.
As to the mechanics of the still-under-revision Artemis Moon mission calendar, there has been some talk of not using the last ICPS upper stage on the Artemis 3 mission so as to save it for later and allow more time for the Centaur-V-for-ICPS swap-out to be engineered and accomplished. If some suitable placeholder spacer can be ginned up to stand-in for ICPS by next year, that certainly looks doable from a vehicle performance standpoint.
I am of the opinion that SpaceX will demonstrate large-scale on-orbit propellant transfer from a tanker Starship to a LEO-orbiting propellant depot Starship before the end of this year. It is even likely that two or more such tanker missions will be run to the same depot ship in 2026.
I am also of the opinion that the initial version of SpaceX’s HLS Starship will be ready to go in time for the Artemis 3 mission at its now-planned mid-2027 target date. Blue Origin’s initial manned lunar lander, I fear, will be a no-show – or perhaps a partially-complete show analogous to the incomplete Orions flown on Artemis 1 and 2. A complete Blue Origin lander of some Mk 1.5 configuration in time for Artemis 3 would be a surprise to me, but a welcome one.
To provide the best possible chance for an early-2028 Artemis 4 initial landing mission to proceed as currently scheduled, it is necessary that SpaceX do its unmanned full-refill, LEO departure, transit, landing, surface loiter and ascent test of HLS Starship as soon as possible. The way to do this, I think, is to keep doing tanker missions to the LEO depot ship until, and even beyond, Artemis 3, as necessary. Then, when an adequately full depot is achieved, refill the HLS Starship lander left in orbit after Artemis 3 and send it Moonward as the lunar landing test article.
Loren Grush recently had a story on Bloomberg News that alleged Artemis 4 might also excuse ICPS from the chore of trans-lunar injection, with the SpaceX HLS Starship docking to Orion in LEO and then proceeding Moonward as a joined unit. Seemingly little-noted was the fact that, if ICPS is relieved of its main function, it might just as easily be relieved of any necessity to show up at all. Thus, we could have two consecutive SLS missions – Artemises 3 and 4, respectively – absent any SLS second stage. This would allow the last ICPS to be used for Artemis V if Blue Origin’s lunar lander is ready – and SLS as well, of course – by late 2028. I do not anticipate that said lander would be able to do the TLI burn in place of ICPS as HLS Starship can.
Under this scenario, we could get all the way through Artemis 5 without needing to fly an SLS with a Centaur V-based upper stage.
As of now there are no more SLSes and Orions mandated for missions beyond Artemis 5. If Artemis 5 slips from its current aspirational launch date of “late 2028,” I can certainly see Isaacman officially playing Taps over both programs before leaving office. That would also mean cancelling the “standardized” SLS with the Centaur V-based upper stage without it ever having flown, but we already have a precedent for that in the form of EUS.
This all assumes – as I very much do – that, by this time, SpaceX will have debuted and tested a crew-capable version of Starship that can take-off from Earth, refill in LEO, then both transit to and from the near-lunar vicinity including a direct re-entry to Earth.
Blue Origin, I expect to also be working on something able to take people to and from the Moon in place of SLS and Orion. Perhaps that will, at least, initially, be Orion riding some version of New Glenn. Isaacman, I’m sure, would be amenable to a stay-of-execution for Orion under such circumstances. That would tide Blue over until it can come up with something better, and far more reusable, that can also fly on New Glenn – probably the 9×4 variant.
Heck, stretching out the executions in this fashion most likely increases the probability that all can be accomplished with the available store of political capital in-hand.
I think Orion/SLS is going to be an exercise in trying to find more delta V somewhere that can be shoehorned into making a lunar mission workable, since the math doesn’t work out with the existing hardware, and attempted architectural fixes like Gateway are massive detours.
As for the broader view of NASA, there are several fundamental changes that hit space flight in quick succession, and I think some of the more important ones stem from SpaceX becoming a commercial success beyond anyone’s imagining. Now there is a launch provider with its own huge and independent revenue stream. Deeply entwined with that success is that it’s very nimble and bold. In competitive terms, it has a very fast OODA loop.
Isaacman has dramatically improved the speed of NASA’s decision making, but he’s got a long way to go and slimming a giant government bureaucracy will not be easy or quick. Even when NASA tries to be bold, that boldness is going to have long timelines because it had to come from long studies, proposals, budget requests, bids, contracts, milestones, design reviews, etc. Basically lots and lots and lots of meetings with lots of officials having to sign off on each step. Meanwhile Elon’s OODA loop is sometimes him just saying “Yeah, lets try that,” and a few months later there’s something new rolling out. And the startups, including many Chinese ones, are trying to follow SpaceX’s approach because smart entrepreneurs emulate success instead of stagnation.
So NASA will likely continue to lag and have some problems making workable plans that interface with the faster, nimbler companies unless they keep old designs around as “legacy hardware” to fulfill NASA requirements from five or six budget cycles earlier. Many of the NASA initiatives will probably have been leapfrogged or bypassed before they are ready to implement.
If that starts proving to be the case, what Isaacman or his successors might have to do is switch to a different, looser contracting model where it defines national goals and priorities that need to be addressed, working with the private sector to see that someone executes the missions. Perhaps a future debate will be why NASA needs to be so big and have so many employees if its role has devolved into setting safety standards and handing out X-Prizes.
You’ve put your finger directly on the crux of the problem in deciding what NASA should be and do, going forward.
Pretty obviously, leadership in the engineering of launch vehicles has already long-since passed to the private sector and the same is in the process of happening to both unmanned and manned spacecraft destined for the Moon and Mars.
I think manned spacecraft for any destination or purpose will be strictly private sector-provided from now on as will an increasing percentage of things like unmanned landers. There will be a lot more of the latter carrying freight than science instrument packages.
As the vast majority of people working in space will have company logos on their coveralls instead of NASA ones within a decade or two, astronaut training will change greatly to cover all of the different types of workers employed extraterrestrially and will be handled by employers, not NASA.
There may remain a few niches for traditional test pilot Right Stuff types, but not many. With future generations of both military and civilian aircraft destined to be unmanned, test pilots, like other types of pilots, will join horse cavalrymen and armored knights in the ranks of formerly essential and valorous vocations that have been rendered obsolete by the relentless march of technology.
NASA will, in pretty short order, be left with only the engineering of pure science missions to the outer planets, the Kuiper Belt and – eventually – the Oort Cloud. And even these projects will increasingly rely on technology developed by the private sector.
NASA seems inevitably destined to shrink quite a bit. Operating test facilities like Stennis and Plum Brook or the wind tunnels at Ames might be viable, longer-term, but even that is far from assured. SpaceX, for example, has its own test facilities at McGregor, TX and has never used Stennis. A number of the newer NewSpace launcher firms still use Stennis, but not even all of them are on-board there.
Plum Brook could also find itself obsoleted. As launch costs diminish, it may soon be cheaper just to put prototype hardware into actual space than to expensively transport it to Ohio and use simulation chambers that are expensive to rent because they are expensive to operate and maintain.
At some point – perhaps as soon as mid-century – it may make more sense to just shutter NASA entirely and transfer whatever science efforts it still pursues to the National Science Foundation.
I agree. I could easily see NASA hewing more to the roll that it had as NACA, and I can see it maintaining a vital role in space-based astronomy and planetary probes (JPL), and perhaps as a major body doing lots of the research on human physiology in space, space medicine, and perhaps as a go-to regulatory authority to try to keep LEO from becoming total chaos.
Basically, cover the needs of government and industry on standards, regulation, and fundamental research, and whatever aspects of science that won’t attract private capital. Formerly, that was pretty much most of space vehicle and space flight research (along with the military) because there was no significant private market until satellite communications and perhaps weather satellites opened one.
One change we might already be seeing is the massive increase in the number of young engineers who are gaining expertise in satellite and space vehicle design and operation. Starlink alone has 3000 employees, and they’ve worked tirelessly to not only design some incredible satellites and constantly improve and rethink them, they’ve also gone beyond that and figured out how to mass produce them commercially, moving well beyond things like assembling bespoke spy satellites hand made by Trappist monks, each with three PhDs. The same is happening at Blue Origin, though in a deep underground bunker somewhere.
And those engineers are also learning or mastering space capsules, space suits, life support systems, control systems, and all the rest. They’ve mastered their own mission control systems, launch systems, and recovery systems.
We’re going to have a space workforce packed with a whole lot of highly capable and experienced young engineers, and just the existence of them should start fertilizing new start ups, similar to what happened in Silicon Valley as all the cool young nerds learned digital electronics and computing. It created a new ecosystem. Perhaps there are parallels to the shift from the mainframe era of IBM, Burroughs, Sperry, Univac, NCR, and Honeywell to what followed, as computing went from a boutique speciality to a ubiquitous everything.
Generally agree, but NASA should not be given any sort of regulatory authority. Particularly not for things like LEO traffic management which is already pretty much a solved problem. It’s been solved by SpaceX and said solution has been offered for free to other satellite constellation operators on a co-operative basis.
The only entities not likely to get onboard are Russia and the PRC. As I don’t see either regime – or either nation, for that matter – lasting much longer, this will be strictly a short-term irritant anent rational space traffic management – especially after every Earth orbit is swept clean of their legacy junk once they’re both gone.
SpaceX has already forgotten more about space traffic management than NASA is ever going to know. And, as there is not a damned thing NASA could do about Russia and the PRC declining to play nicely with others in orbit, there is absolutely no reason to charge it with regulating the spaceways. NASA simply isn’t up to the job.
SpaceX having a glory-days-of-Silicon-Valley affect on the aerospace industry generally is already very much a thing. SpaceX alumni can be found everywhere in more recently founded NewSpace companies – many in founder and senior management roles. The size of the “SpaceX Diaspora” powering all of this ferment will only increase as the company’s workforce continues to expand.
DE writes:
” Where his piece falls short is in its vision of the future. It’s still a NASA-centric one.”
NASA has not made the mental leap to moving from service provider to service consumer.
And even Isaacman encourages being stuck in the past when he says he wants to re-acquire that Good Old Expertise.
I’m not sure if Isaacman really believes that or if he is just saying that to mollify the masses as he puts his changes in place. I’m hoping for the latter and I do see some indication that he is going to make the switch. e.g. Gateway Pause.
That crew and capsule recovery evolution was embarrassing. Only because I witnessed it can I imagine ways to have made it more complex. But if I had to design it, I don’t think I could have found as many ways to waste time and money as NASA accomplished in the 2+ hours it took to just extract the crew.
One offshore installation vessel with a 100 ton crane, and they would have had the capsule and crew on a dry deck in 30 min, and then aboard 1 helicopter for the flight back to land within those 2 hours.
Yeah, that was embarrassing to watch. Not a good-looking way to finish the show. If Artemis 2 can be compared to a vaudeville act, that Keystone Kops recovery display would be the rough equivalent of the Irish Tenor’s pants falling down as he’s taking his bow.
Your description of the reasonable alternative is spot-on and could be shortened to just – “do it the way SpaceX does it.”
Obviously NASA didn’t rethink things during the years since Apollo, and just dusted off the old 1960’s documentation on how to recover a capsule after splashdown.
So perhaps it’s surprising that NASA used the Murtha, LPD-26 an amphibious transport dock, instead of the USS Hornet, CVS-10, which recovered Apollo 11 and 12, which they left sitting idle in Alameda California. The Intrepid and Yorktown are on the east coast, and could be made available for Atlantic Artemis splashdowns. Real space missions use Essex class carriers.
As I watched the recovery once the helicopters were on board the ship, I noticed that NASA only had one guy who had to escort each astronaut from a helicopter to a hatch, and then he’d have to go back to the helicopter to get another astronaut, and another. Obviously NASA needs to employ three additional escorts to speed things along. And why wasn’t anyone playing trumpets or throwing confetti? The recovery was badly understaffed, which wouldn’t be the case if they’d use an Essex class carrier with 2600 men, instead of just 550 people on an LPD. So what if SpaceX’s recovery ship only uses forty people?
Anyway, it’s odd that it takes so many experts and medical specialists to get astronauts out of a capsule during a normal recovery, yet if the capsule makes an emergency splashdown the astronauts are good to climb out of the capsule on their own, swim to an island, and live by hunting marine reptiles with a spear for a couple of weeks until they’re located. I’m sure that if left on their own, they’d eventually motivate and start doing things themselves.
Heck, now that Starlink is moving into providing global phone service, maybe we could just have the capsule come down where ever, and then the astronauts go online, call for an Uber, and make arrangements with a shipping company to freight the capsule back to Florida, so that the recovery force doesn’t actually employ anybody.
Heh.
Some more thoughts I had.
Given that three of the crew were F-18 pilots, it occurs to me that F-18 pilots routinely pull more G’s than the Orion’s max of 3.9 G’s, eyeballs in, and F-18 pilots go through far more violent maneuvers, and almost random ones at that. High G’s, shaking, bouncing, getting slammed around in the cockpit, negative G’s, high spin rates, and varying cockpit air pressures. Heck, during the arrested landing, an F-18 pilot goes through more G’s eyeball-out than an Artemis crew feels eyeballs-in.
So what if the Navy did F-18 recoveries like NASA does Artemis recoveries? First they’d let the pilot sit in the cockpit for half an hour or so, then send in a medical specialist to check him out. Then a team of medical specialists would lift him out of the cockpit and gently lower him to the deck, and then sit with him for half an hour before escorting him to sick bay for further tests.
But after a century of naval aviation, this would strike everyone as absurd. We know the fighter pilots are fine after recovery, even if their teeth might have gotten a little rattled.
With the first astronauts it made sense to have doctors rush in to get the freshest possible blood samples to see if weightlessness had any weird effects on them, and run them through batteries of tests to look for changes induced by their longer and longer stays in zero-G. But we’ve now recovered over 700 people from space missions, and not a single one has just keeled over dead after landing. They’ve all come through fine. We don’t need to keep treating them like the Mercury Seven or Gemini crews. We don’t need to keep treating them like lab rats who might hold the key to understanding space flight, and we don’t need to give each one a ticker tape parade through Times Square, either. Just because we started out doing it doesn’t mean we have to keep doing it.
Agree again, though the percentage of space travelers with air combat maneuvering and carrier take-offs and landings on their resumes will dwindle sharply in coming years but that will be compensated for by gentler rides both up and down.
A Starship will subject its passengers to no more G-forces on ascent than those experienced by today’s Orion, Starliner and Dragon riders. The ride uphill should also be notably smoother as Starship has no giant flaming organ pipes attached to rattle passenger teeth and it has way more engines on each stage than a Falcon 9. On both stages, said engines are also much farther away from the passenger accommodations.
On return, Starship subjects its passengers to no more G-force than Shuttle. The only thing even faintly alarming might be the flip-up maneuver prior to chopstick catchment and this will be mild even compared to typical amusement park rides.
Anent ticker tape parades, those are no longer possible due to both the modern equivalent of stock tickers – Bloomberg terminals – producing no paper tapes and to the fact that windows in modern high-rise buildings don’t open to allow any notional ticker tape or other forms of paper to be thrown.
What about using drones to drop CVS receipts?
Heh again.