Commercial LEO Destinations Development

[jongoff]

Sounds like total reset of the requirements, much more relaxed now. Not too sure what "4 crew for 1-month increments" entails, 1 month mission is ok? Does the next crew needs to take over after 1 month or can the station be unmanned for a while? I think Vast's Heaven-1 already plan to support 4 crew for one month, wonder if this is what Duffy has in mind.

My interpretation is that they're open to a crew-tended 4-person station that can handle crew on-board for at least 1 month stints at a time, but could then be uncrewed for periods of time between missions. Basically, meaning that Vast's Haven-1 would be compliant with the minimum expectations.

So yeah, if any of the CLD players were counting on Vast not being able to compete due to requirements, that hope is now gone. But it's worth mentioning they were also quite explicit about wanting to keep 2+ players involved through this continued SAA phase, so even though Vast is likely to now have a good shot at getting CLD money, there's still room for at least 1-2 others.

I think these changes are very good. Pushing toward an operational capability that can be certified later I think makes a lot of sense for moving faster and at lower cost (better aligned with budget availability between now and when ISS retires). There may be some negative nuances I'm missing, but I think this is a pretty positive change overall.

~Jon

[Vultur]

From Berger's article: https://arstechnica.com/space/2025/08/as-the-end-of-the-iss-nears-nasa-shakes-up-program-for-commercial-replacements/

"All the current players are going to have to do some kind of pivot, at least revisit their current configuration," McAlister said. "Certain players are going to have to do a harder pivot."

One industry official, speaking anonymously, put it more bluntly: "Only Haven-1 can succeed in this environment. That is our read."

The chief executive officer of Vast, Max Haot, told Ars that the company bet that starting with a minimum viable product was the best business strategy and fully funded that approach without government money.

"It seems like NASA is now leaning into an approach for the future of CLDs that is led by what industry believes it can achieve technically and build a credible business model around," Haot told Ars. "Seeing that information from contractors before committing to buying services can help increase long-term risk reduction. This seems similar to the successful approaches used by NASA for cargo and crew."

I think I agree with Haot, this is a very good move by Duffy. The original CLD requirement is stupid, it tries to gold plate the new station and make it as close to ISS as possible, without examining why do we even need that. This resulted in all the awardees designing battlestar galactica stations that have no chance of self-sustaining and would mostly depends on NASA funding.

The pivot to minimal requirement and MVP aligns this program with the best practices of commercial space.

What requirements does such an 'MVP' actually deliver on? Can it host the long-term internally-maintained and externally exposed instruments and experiments that have defined the ISS' scientific output? Can it host the long-duration human missions that the ISS allows for biological studies (e.g. the twins studies)? etc.
An ISS replacement should be an improvement on existing capabilities - or at the absolute least maintain those capabilities - not a re-run of the Salyut programme from half a century ago.

I'm not completely convinced (at least, if you mean "should" from the US government's perspective, as opposed to the space life sciences community's perspective or your/my personal perspective).

I think it is arguable that the science return of ISS does not necessarily justify its cost today, at least to the US government.

Its geopolitical goal (which I would argue was probably a more important reason it both existed and has survived this long) is much less relevant in today's environment (the equivalent would have to be a US-China combined project, which neither country is likely to do).

So it is possible that the real purpose for this is just for the US to show we're not abandoning LEO, giving the US "a seat at the table", etc.

From that perspective, science capabilities are secondary (as I'd argue they usually are for human spaceflight).

[thespacecow]

What requirements does such an 'MVP' actually deliver on? Can it host the long-term internally-maintained and externally exposed instruments and experiments that have defined the ISS' scientific output? Can it host the long-duration human missions that the ISS allows for biological studies (e.g. the twins studies)? etc.
An ISS replacement should be an improvement on existing capabilities - or at the absolute least maintain those capabilities - not a re-run of the Salyut programme from half a century ago.
CCDev did not restart commercial space programmes with suborbital crew hops, after all: the minimum requirement was the operational goal.

Commercial Crew *is* the equivalent of rerun of Salyut program, they didn't try to improve or even maintain the capabilities of Shuttle: 7 crew, large cargo mass/volume, robotic arm, 2 weeks in orbit, etc. None of these are met by CC vehicles, which in fact caused the US government to lose some capabilities, most notable the ability to service Hubble. In the end CC vehicles are much closer to Apollo capsules than to Shuttle, I believe this was explicitly commented on in the early days.

But there's nothing wrong with that. ISS like Shuttle is a large government program which have tons of gold plate requirements, a lot of that is not cost effective or commercially viable. Returning to the basics is what's needed to figure out a commercially sustainable pathway forward. Once the new players have more experience, they can iterate back towards more capability on their own terms, just like post Commercial Crew we now have a vehicle much more capable than Shuttle being developed by a company.

These new requirements smacks of picking whatever commercial station was closest to flying, and then defining the programme requirements from that so you can get an easy 'win'. That's the same thinking that gave us Constellation and SLS.

"Picking whatever was closest to flying" is not at all what Congress did wrt SLS, if they actually did that they'd picked D4H and FH instead of SLS. But instead Congress made requirement skyhigh, 130t to LEO, because "we just had to have this superheavy capability, existing commercial capability is just not enough, because reasons", sounds familiar? The requirement goes well beyond what commercial LV can do at the time, that's why we end up with the boondoggle like SLS.

[woods170]

It's worth noting that Axiom and others have testified before a congressional committee on this topic not too long ago.
Dittmar's testimony is linked here as just one example.
https://democrats-science.house.gov/imo/media/doc/Dr.%20Dittmar%20-%20Testimony.pdf

🤮🤮🤮

Typical for Dittmar:

"NASA is wasting money on companies it will never select."

Ahem, that's not a bug. Having multiple contenders at first, before downselecting to just one or two later on, is a feature of most NASA competitions. NASA's been doing this since (at least) the Space Shuttle days and did it for Space Station Freedom (which became the ISS) as weel. They also did it for CRS and CCP. Most recently this way of doing things was applied to HLS. And now CLD.

Mary Lynne knows this all too well, but she deliberately tries to change the narrative into it being something bad.

If it were up to Mary Lynne NASA would have selected Axiom immediately as the sole contractor for CLD. She's been standing on that particular soapbox for years now. It's becoming annoying. Much like Dittmar was darn annoying when she was still the President and CEO of the Coalition for Deep Space Exploration. In that capacity she acted much like David Willis does on Twitter: cheerleading for SLS despite everyone correctly and justifiably pointing out how flawed and unaffordable the darn thing is.

[woods170]

What requirements does such an 'MVP' actually deliver on? Can it host the long-term internally-maintained and externally exposed instruments and experiments that have defined the ISS' scientific output? Can it host the long-duration human missions that the ISS allows for biological studies (e.g. the twins studies)? etc.
An ISS replacement should be an improvement on existing capabilities - or at the absolute least maintain those capabilities - not a re-run of the Salyut programme from half a century ago.
CCDev did not restart commercial space programmes with suborbital crew hops, after all: the minimum requirement was the operational goal.

Commercial Crew *is* the equivalent of rerun of Salyut program, they didn't try to improve or even maintain the capabilities of Shuttle: 7 crew, large cargo mass/volume, robotic arm, 2 weeks in orbit, etc. None of these are met by CC vehicles, which in fact caused the US government to lose some capabilities, most notable the ability to service Hubble. In the end CC vehicles are much closer to Apollo capsules than to Shuttle, I believe this was explicitly commented on in the early days.

You seem to be forgetting the history behind the Space Shuttle that caused this to happen: Both the "Presidental Commission on the Space Shuttle Accident" (Challenger) and the "Columbia Accident Investigation Board" (Columbia) noted that the function of the Space Shuttle being two-fold (flying both humans AND substantial amounts of cargo on the same vehicle) severely increased the risk to the crew. The "substantial cargo" requirement is what led to Shuttle becoming a winged vehicle requiring a fragile, yet reusable, heat shield and massive boosters to get the thing out of the lower atmosphere.
Guess what killed two missions: the fragile heat shield and the massive boosters.

So, when a replacement of the Space Shuttle was needed, NASA justifiably did away with the requirement to fly substantial amounts of cargo on the crewed vehicle. That served to do away with the need for "wings into space" and thus the demise of space planes.

Lest you forget: the CCP vehicles ARE an improvement over the Space Shuttle: the crew now has abort capabilities (and thus the chance to survive a launch mishap) all the way into orbit. They also have a much better chance of surviving reenty, now that they fly in a vehicle who's natural COG points the heat shield automatically in the correct direction. AND has a much more reliable and sturdy heat shield to begin with.

Seven crew up capabilities were never required for the role the Space Shuttle was originally assigned: to construct a space station. Several of the ISS construction missions were flown with crews of only 5 astronauts. And for missions flying to a completed space station, having 7-person crews make no sense whatsoever. Which is why the requirements for CCP were flying four crew members.

[DanClemmensen]

Seven crew up capabilities were never required for the role the Space Shuttle was originally assigned: to construct a space station. Several of the ISS construction missions were flown with crews of only 5 astronauts. And for missions flying to a completed space station, having 7-person crews make no sense whatsoever. Which is why the requirements for CCP were flying four crew members.

Both Crew Dragon and Starliner were designed for a crew of seven. For Dragon, NASA and SpaceX shifted from retro-propulsive landing to parachute splashdown to shorten the schedule by using the simpler and better-understood system, but this required seats that could tilt, to the "parachute landing" configuration. and that takes up too much space. Nevertheless, NASA/SpaceX still believe that Dragon can land with up to seven  crew in a rescue situation. Launch with seven is not actually a problem.

[yg1968]

New article...

https://arstechnica.com/space/2025/08/as-the-end-of-the-iss-nears-nasa-shakes-up-program-for-commercial-replacements/

Interesting that the crewed demo mission would involve non-NASA crew. From the Directive that is linked in that ARS Technica article:

• SAA scope should include milestones maturing CLD systems to CDR readiness and then an initial in-space crewed demonstration. Flexibility should be granted to industry regarding the extent of demonstration capability (this in-space crewed demonstration would be non-NASA crew).

https://cdn.arstechnica.net/wp-content/uploads/2025/08/NASA_CLD_Directive_Aug_4-1.pdf

[StraumliBlight]

Pam Melroy wrote an Op-ed for Payload: NASA’s New CLD Strategy Will Lose Mars, LEO to China [Aug 8]

Mission shift: While it may be possible to evolve a commercial space station to support longer-duration stays for future missions, not requiring that it be considered in the design up front historically means that NASA will pay more for transportation in the near term, then pay again to upgrade to long-duration flight. I can visualize the headlines in five years when this type of short-term decision-making bears ugly cost and programmatic consequences, and national capability implications. And NASA will be rightly criticized for doing it to themselves.

Another problem is changing the strategy in the 11th hour for industry—never a good idea during acquisition. Industry, and its investors, have anticipated NASA will start the competition for commercial LEO destinations in earnest this year. There is a real question about how much capital can be raised for a private space station, and this change will not make it easier.

Many aerospace industry partners clearly signaled to NASA that their funding sources desired a final competition, guaranteeing return on investment to the funders in the next phase. Extending development instead of going to final competition, as required by Duffy’s new directive, will likely save NASA some money up front, but reduce competition—ultimately driving up costs and hurting innovation—as some industry partners drop out due to lack of funds.

On this episode of Valley of Depth, we’re joined by Pam to talk about the state of NASA, the future of space stations, and why requirements, the often overlooked backbone of program management, will determine whether the U.S. stays ahead. We trace her career from test pilot to shuttle commander to senior leadership at NASA, DARPA, and the FAA, and unpack what it means to build an architecture that actually holds together from LEO to Mars.

[sdsds]

Pam Melroy wrote an Op-ed for Payload: NASA’s New CLD Strategy Will Lose Mars, LEO to China [Aug 8]

Many aerospace industry partners clearly signaled to NASA that their funding sources desired a final competition, guaranteeing return on investment to the funders in the next phase.

This is the key observation: private capital markets disfavor (charge for) uncertainty. This move thus increases the cost of capital for CLD ventures.

[clongton]

Seven crew up capabilities were never required for the role the Space Shuttle was originally assigned: to construct a space station. Several of the ISS construction missions were flown with crews of only 5 astronauts. And for missions flying to a completed space station, having 7-person crews make no sense whatsoever. Which is why the requirements for CCP were flying four crew members.

Speaking strictly of NASA missions, there are no destinations anywhere in LEO, cis-lunar space or on the lunar surface that require more than a 4-crew spacecraft. And it will stay that way for a very long time. On the rare occasion where a mission might benefit from more crew for some reason, SpaceX can simply launch a second spacecraft. The cost, in relative terms to what it used to cost NASA for just 1 old-space launch, is a mere pittance, lost in the margins. Over time that will definitely change, but not in my lifetime. The HLS might moderate that slightly, but only if it's a pure SpaceX mission, not NASA. NASA won't need more than 4 crew at a time for a very, very long time. Congress will never pony up what NASA and its buddy contractors would demand for that. Nope. For NASA, the massive capability of the HLS will be cargo delivery, not crew.

[clongton]

Both Crew Dragon and Starliner were [was] designed for a crew of seven. For Dragon, NASA and SpaceX shifted from retro-propulsive landing to parachute splashdown to shorten the schedule by using the simpler and better-understood system, but this required seats that could tilt, to the "parachute landing" configuration. and that takes up too much space. (...)

The switch was made because NASA was risk averse and unwilling to employ a NIH propulsive landing system, not to shorten the schedule. This was covered ad nauseum in previous threads years ago.

Edit: Here is the presentation, including a CGI. It was the Dragon V2 unveil event on May 29, 2014. EDL begins around 5:15 and propulsive landing is at ~6:12. Mr. Musk stated at 6:31 "THAT is how a 21st century spaceship is supposed to land". His statement was emphatic and that is exactly how he intended Dragon to land, until NASA told him otherwise. At that point SpaceX still needed NASA to survive. I seriously doubt that today he would be so accommodating.

[thespacecow]

So, when a replacement of the Space Shuttle was needed, NASA justifiably did away with the requirement to fly substantial amounts of cargo on the crewed vehicle. That served to do away with the need for "wings into space" and thus the demise of space planes.

The Commercial Cargo vehicle's up/down mass capacity doesn't match Shuttle's either, so even if you take the crew/cargo split into account, the capability regression from Shuttle is still there.

Lest you forget: the CCP vehicles ARE an improvement over the Space Shuttle: the crew now has abort capabilities (and thus the chance to survive a launch mishap) all the way into orbit. They also have a much better chance of surviving reenty, now that they fly in a vehicle who's natural COG points the heat shield automatically in the correct direction. AND has a much more reliable and sturdy heat shield to begin with.

Well yeah, one would hope 50 years later we have made some improvement... Obviously the new stations would have some improvements over ISS too, for example they wouldn't cost so many crew hours to maintain, but that's not the point.

The point is not trying to match old system's capability one by one, instead think critically about each requirement, eliminate dumb requirements and only keep those absolutely necessary.

Seven crew up capabilities were never required for the role the Space Shuttle was originally assigned

This is the point, not all the capabilities of the old system is worth preserving.

[sdsds]

This is the point, not all the capabilities of the old system is worth preserving.

As another example of this, none of the CLD station designs support delivery of an entire International Standard Payload Rack, since none of them offer CBM-sized access hatches. Everything moving in or out needs to fit through an 800 mm diameter docking port. (At least I'm unaware of any CLD stations offering CBM. Perhaps that's merely a sign of ignorance, though!)

[edzieba]

Both Crew Dragon and Starliner were [was] designed for a crew of seven. For Dragon, NASA and SpaceX shifted from retro-propulsive landing to parachute splashdown to shorten the schedule by using the simpler and better-understood system, but this required seats that could tilt, to the "parachute landing" configuration. and that takes up too much space. (...)

The switch was made because NASA was risk averse and unwilling to employ a NIH propulsive landing system, not to shorten the schedule. This was covered ad nauseum in previous threads years ago.

That's not quite what happened.
When CCDev was being drawn up, the assumption was that SpaceX would self-fund the testing to qualify propulsive landing (just as they would self-fund other developments, like the abort system, their parachutes, and their own IDSS-compliant docking system), and gain back tyhose development costs over the overall programme profit for operating Dragon for the years to come. SpaceX were happy to do this, as it would also act as sub-scale testing for BFR, which at the time was a base-first capsule.
As BFR development progressed, BFR switched to the side-first 'skydiver' entry, and Dragon propulsive landing was now a cost purely to the Dragon programme alone rather than spread over multiple programmes. SpaceX no longer wanted to fund a dead-end tech themselves, so switched to parachutes, which NASA were happy to agree to as parachutes were well understood* and NASA could provide plenty of development support.
This is the same thing that killed Red Dragon: Red Dragon would have been a money sink of a Mars EDL subscale demonstrator for BFR development, but that could also claw back a bit of development budget (and get some early external customer buy-in for Mars transport) by carrying some actual payloads rather than just a mass simulator. When the need for a sub-scale base-first Mars retropropulsion demo vanished with the BFR design change, so too did the SpaceX internal need for Red Dragon - and there was no way to keep it alive by asking the prospective payloads (who were previously getting a ride for a song) to suddenly bear the full development and operational cost of the entire programme.


*It turned out in the end that parachutes were not as well understood as everyone thought (even NASA), but SpaceX would have had to know that in advance to influence the decision to to drop Dragon propulsive landing many years before, and even then parachutes would probably still have traded as cheaper.

[jarmumd]

This is the point, not all the capabilities of the old system is worth preserving.

As another example of this, none of the CLD station designs support delivery of an entire International Standard Payload Rack, since none of them offer CBM-sized access hatches. Everything moving in or out needs to fit through an 800 mm diameter docking port. (At least I'm unaware of any CLD stations offering CBM. Perhaps that's merely a sign of ignorance, though!)

I'm going to get on my soapbox a bit, because to me, this is a clear example of where "commercial" breaks down.  What I mean is, if it were pure commercial, there would be an independent org like IEEE, SAE, etc where the business community would support the development of the "next gen" port (think usb/thunderbolt/wifi).  That doesn't exist today because there is too much risk (ie dev money) in creating a new standard.  Furthermore, since IDSS is quasi open source, at least you can kinda build to it.  AFAIK CBM is proprietary - well I only know of folks building passive CBM, I don't know of any active CBM other than what's on ISS.

I think CLD would love a larger port than IDSS.  But the reality is that there isn't money to support an alternative standard that at first glance we don't need (assume everything can move through the Apollo sized 800mm IDSS inner diameter).

[clongton]

Both Crew Dragon and Starliner were [was] designed for a crew of seven. For Dragon, NASA and SpaceX shifted from retro-propulsive landing to parachute splashdown to shorten the schedule by using the simpler and better-understood system, but this required seats that could tilt, to the "parachute landing" configuration. and that takes up too much space. (...)

The switch was made because NASA was risk averse and unwilling to employ a NIH propulsive landing system, not to shorten the schedule. This was covered ad nauseum in previous threads years ago.

That's not quite what happened.
When CCDev was being drawn up, the assumption was that SpaceX would self-fund the testing to qualify propulsive landing (just as they would self-fund other developments, like the abort system, their parachutes, and their own IDSS-compliant docking system), and gain back tyhose development costs over the overall programme profit for operating Dragon for the years to come. SpaceX were happy to do this, as it would also act as sub-scale testing for BFR, which at the time was a base-first capsule.
As BFR development progressed, BFR switched to the side-first 'skydiver' entry, and Dragon propulsive landing was now a cost purely to the Dragon programme alone rather than spread over multiple programmes. SpaceX no longer wanted to fund a dead-end tech themselves, so switched to parachutes, which NASA were happy to agree to as parachutes were well understood* and NASA could provide plenty of development support.
This is the same thing that killed Red Dragon: Red Dragon would have been a money sink of a Mars EDL subscale demonstrator for BFR development, but that could also claw back a bit of development budget (and get some early external customer buy-in for Mars transport) by carrying some actual payloads rather than just a mass simulator. When the need for a sub-scale base-first Mars retropropulsion demo vanished with the BFR design change, so too did the SpaceX internal need for Red Dragon - and there was no way to keep it alive by asking the prospective payloads (who were previously getting a ride for a song) to suddenly bear the full development and operational cost of the entire programme.


*It turned out in the end that parachutes were not as well understood as everyone thought (even NASA), but SpaceX would have had to know that in advance to influence the decision to to drop Dragon propulsive landing many years before, and even then parachutes would probably still have traded as cheaper.

I updated my post above [#510] to provide context

[woods170]

So, when a replacement of the Space Shuttle was needed, NASA justifiably did away with the requirement to fly substantial amounts of cargo on the crewed vehicle. That served to do away with the need for "wings into space" and thus the demise of space planes.

The Commercial Cargo vehicle's up/down mass capacity doesn't match Shuttle's either, so even if you take the crew/cargo split into account, the capability regression from Shuttle is still there.

Wrong. The cargo requirements laid down by NASA for CRS were not even close to those laid down for STS in the 1970s. The only reason Shuttle had an 18 meter long cargo bay and a capability to bring up 27 metric tons to LEO was the Space Station and certain military payloads. The military requirements for NASA vehicles went away permanently in the early 2000s and are no longer a factor to be considered.

CRS didn't have to carry up 27 metric tons of cargo, but only 15% of that (at most). With the ISS complete NASA no longer has a requirement to carry 27 metric tons to orbit on one launch. Let alone a requirement that needs an 18 meter long cargo bay.

As such, the CRS vehicles are not regression, because the customer doesn't require shuttle-sized numbers anymore. It makes no sense to build a cargo vehicle (Cargo Dragon) that has only one customer (NASA) and then give that vehicle six times the required cargo capacity. Even SpaceX doesn't do that.
And before you begin about Starship: NASA is not its only customer. The over-sized specs for Starship come from SpaceX's own needs, not NASA's.

[pathfinder_01]

I'm going to get on my soapbox a bit, because to me, this is a clear example of where "commercial" breaks down.  What I mean is, if it were pure commercial, there would be an independent org like IEEE, SAE, etc where the business community would support the development of the "next gen" port (think usb/thunderbolt/wifi).  That doesn't exist today because there is too much risk (ie dev money) in creating a new standard.  Furthermore, since IDSS is quasi open source, at least you can kinda build to it.  AFAIK CBM is proprietary - well I only know of folks building passive CBM, I don't know of any active CBM other than what's on ISS.

I think CLD would love a larger port than IDSS.  But the reality is that there isn't money to support an alternative standard that at first glance we don't need (assume everything can move through the Apollo sized 800mm IDSS inner diameter).

I would say that not developing a larger port at this time is very much a good idea. The only vehicles that could deliver the racks were the Shuttle and the Japanese HTV. This is where commercial shines at eliminating possibly unneeded things. The CBM. What you listed were just organizations that develop standards. Companies are free to use what ever standards to do anything. They are only limited by contracts and by law.   CLD might like a larger port but it did not like it enough to specify that it be that large, be CBM or be CBM compatible.

[sdsds]

Is it correct that Axiom Station is already somewhat 'bought in' to the CBM standard?

[jarmumd]

Is it correct that Axiom Station is already somewhat 'bought in' to the CBM standard?

I believe that's correct at least insofar as connecting to the ISS.  Their graphics show CBM with IDSS adapter ports.

And I do agree that if nothing needs the larger port, than it's an unnecessary requirement.  What I caution, is that we may be applying an operations logic to a space station construction, where we might need outfitting with large racks.  Unclear if we can launch space stations fully outfitted.