Alternatives to Current Artemis Program Architecture Discussion Thread

[jadebenn]

I humbly suggest this thread as a more constructive alternative than having these discussions in the main Artemis thread.

[DanClemmensen]

I humbly suggest this thread as a more constructive alternative than having these discussions in the main Artemis thread.

This looks good to me, for the purpose of discussing details of the alternatives. However, the fact that a viable alternative might exist will have consequences for the current plan of record, and I feel these consequences can and should be discussed in the main thread. Example: discuss the details and timing of a purely Starship-based crew transport here, but mention the potential funding impact in the main thread if appropriate. Thus, the existence of this "alternatives" thread should not be used as an excuse to shut off discussion of those consequences in the main thread.

[spacenut]

Upgrade Dragon II and Falcon Heavy can launch it to the moon and back.  The Dragon trunk can contain a small service module with rocket engine and fuel if needed. 

Falcon Heavy can also launch all the modules for the Artemis station. 

Without Starship, a lunar lander could be developed to fit Falcon Heavy that could also go to the Artemis Station.  Fuel can also be delivered.

With Falcon Heavy at a base price of $85 million, it could launch 23 launches for the cost of one SLS. 

Then when Vulcan Heavy and New Glenn come on line, SLS becomes unnecessary due to COST alone.  Also the overweight Orion capsule is not needed. 

This is not even counting the Starship/Superheavy system. 

More launches, more rocket companies can be involved, thus money is spread around if all of SLS money is spent using distributed launches for everyone and keeping all components under 20 tons.  ISS was built like this. 

[TrevorMonty]

For crew transport between from earth to Gateway and back I prefer OTV. This would do LEO-Gateway-LEO trip and refuelling at Gateway for return leg. At 14.5t wet and 5t dry includes payload ie crew, Methalox OTV has enough DV for one leg. Could be delivered to Gateway fully fuelled by Vulcan Centaur.
Scenario 1.
1) Crew fly LEO on Dragon, transfer to OTV.
2) Another RLV (Neutron, SS, F9R, take your pick) delivers 9.5t of fuel to refuel OTV in LEO.
3) Vulcan Centaur launches with no payload and should arrive in LEO with 27t of residual fuel. It then docks with OTV in LEO,        then delivers it directly to Gateway.
4) OTV returns to LEO at end of mission and delivers crew to waiting Dragon.

Scenario 2
1 & 2 are same.
3) OTV flys it's self to Gateway.
4) Refuels at Gateway and returns to LEO.
SS or Vulcan can be used to deliver return fuel to Gateway.

Scenario 1 has lower risk as most of high risk dockings and refueling maneuvers happen in LEO. Crew could wait until fully fuelled OTV is docked to Centaur before boarding.

[deltaV]

This post is about the procurement strategies NASA should use to launch large payloads if SLS is canceled or a redundant launch vehicle is desired. I’m taking it for granted that fixed price contracting and continuing competition should be used. I see two broad strategies NASA could take: purchasing launch services directly, or purchasing launch only indirectly as a part of an end-to-end service such as moving people from Earth to the moon and back. As I’ll discuss in more detail at the end of this post I like the indirect approach better than the direct approach because it offers more opportunities for creative bids to save NASA money. In both approaches NASA should design the requirements flexibly to allow a variety of architectures, especially multiple launches of smaller launch vehicles using propellant transfer or in-space assembly.

In the direct purchase approach NASA would make standardized requirements for all large NASA launches analogous to what the EELV program did for DOD payloads. A standard mission could be launching an X tonne payload towards the moon. NASA should start by asking both NASA launch consumers and potential launch vehicle providers what values of X would be convenient for them. I bet X around 45 tonnes would be a good compromise, which would presumably allow Starship, New Glenn, and a future three core Vulcan to compete (using multiple launches and propellant transfer or in-space assembly). NASA would do a bulk buy of launch services from two providers meeting the requirements (like the EELV program did), paying a portion of the development cost if needed. NASA programs would design to the same standardized requirements.

In the indirect purchase approach NASA would purchase end-to-end goods and services such as moving people from Earth to the moon and back or building a surface habitat and delivering it to the moon. The bidders would be responsible for subcontracting with launch service providers of their choice. NASA should increase flexibility by accepting trade-offs between number of missions and capability, e.g. allow 10 moon missions with 2 astronauts each or 5 missions with 4 astronauts each. This would allow bidders using vehicles as small as Terran R (20 tonnes to LEO) to submit bids using multiple launches and propellant transfer or in-space assembly, though bidders using small launch vehicles could have trouble competing on price given the large number of launches they'd need to do.

In the indirect purchase approach there’s a question of what, if anything, to do to ensure the providers use different launch vehicles for redundancy and competition. I think NASA should avoid having both providers depend on any major component (such as a rocket or rocket stage) that has not yet flown successfully but allow both to use components that have already flown. I think having both providers use an already-operational rocket is acceptable for a couple of reasons. Firstly major delays of already-flown rockets tend to be limited to accident investigations and stopping moon missions for a year or so occasionally during an investigation isn’t a big deal compared to the billions of dollars that redundant launch vehicles could cost. Secondly the rockets serving NASA would probably have other customers so all NASA human spaceflight missions using one launch vehicle provider would be unlikely to drive everyone else out of business. Thirdly even if a monopoly develops in super heavy launch vehicles (e.g. Starship) the threat of the prime contractors using a smaller launch vehicle or developing a new super heavy vehicle would force the monopolist to perform.

I like the indirect approach better than the direct approach because it offers more opportunities for creative bids to save NASA money. One example is SpaceX’s unexpected use of Starship as a lunar lander was possible with the HLS program, which took the indirect approach, but wouldn’t have been possible with the direct approach since Starship is way too big to launch as a standard payload. Another example is the indirect approach easily allows the use of a solar-electric tug to transfer cargo from LEO to lunar orbit whereas the direct approach could only incorporate a tug if NASA designed the requirements specifically for that.

What do you all think?

[yg1968]

At 44 and 45 minutes of this video, Jim Bridenstine talked about the alternative of using Orion on a FH with an ICPS. He said that it wouldn't accelerate the (Artemis I) 2020 mission but it could retire risk for the 2024 (Artemis III) mission. He spoke of an all of the above strategy which would presumably include SLS/Orion and the FH/Orion commercial alternative.



Here is the transcript of the conversation:

[Mr. Bridenstine]So then we looked at way out of the box--what if we were to consider putting a Falcon Heavy with an Orion service module or the Orion crew vehicle and a European Service Module and an ICPS from ULA? I know that sounds crazy, but again, we're looking at all options. And in fact it works. It requires a lot of modifications to the launch infrastructure, to the launchpad, to the erector arm. It takes a lot of modifications
to do cryogenic and hypergolic refueling on the pad, which doesn't currently exist. There--it takes a lot of time and there's a lot of cost and there's risk, and it wouldn't work for accelerating a 2020 launch of an Orion crew vehicle.

    But what it did demonstrate is that if you have a little bit of extra time, 2023, maybe 2024, a lot of that uncertainty
could be retired. And if we're going to get----

    Ms. Horn. So just because I have a few more questions----
    Mr. Bridenstine. Yes.
    Ms. Horn [continuing]. That I want to get to. So can you boil it down to the final decision? Because I appreciate you
looking at all those things, but can you boil that down to the final decision that we're still on track----
    Mr. Bridenstine. The----
    Ms. Horn [continuing]. With the SLS for EM-1?
    Mr. Bridenstine. Absolutely.
    Ms. Horn. OK.
    Mr. Bridenstine. SLS is the best--in fact it's the only option for EM-1, and there are options in the future that need
to be considered. And when we land on the moon in 2024, it's only----
    Ms. Horn. OK.
    Mr. Bridenstine [continuing]. Because of an all-of-the-above strategy.

https://www.govinfo.gov/content/pkg/CHRG-116hhrg35788/html/CHRG-116hhrg35788.htm

Pence a couple of weeks earlier had also mentioned the possibility of using commercial rockets for the (Artemis III) 2024 mission.

And if commercial rockets are the only way to get American astronauts to the Moon in the next five years, then commercial rockets it will be.

https://trumpwhitehouse.archives.gov/briefings-statements/remarks-vice-president-pence-fifth-meeting-national-space-council-huntsville-al/

[punder]

At 44 and 45 minutes of this video, Jim Bridenstine talked about the alternative of using Orion on a FH with an ICPS.
…

Bridenstine:…Falcon Heavy…
Horn: Shut up. Now: Repeat after me, SLS is awesome. Got it? Good.

[Timber Micka]

Falcon Heavy can also launch all the modules for the Artemis station. 

They need a service module to enter lunar orbit and rendez-vous with the PPE/HALO stack in NRHO on their own. They cannot do that with their current design, which is frozen and it is too late to change it since they are already under construction at Thales Alenia.

Without Starship, a lunar lander could be developed to fit Falcon Heavy that could also go to the Artemis Station.  Fuel can also be delivered.

That would end up being a 3-segment lander like the National Team proposal, with the same flaws and shortcomings.

[TheRadicalModerate]

Falcon Heavy can also launch all the modules for the Artemis station. 

They need a service module to enter lunar orbit and rendez-vous with the PPE/HALO stack in NRHO on their own. They cannot do that with their current design, which is frozen and it is too late to change it since they are already under construction at Thales Alenia.

A fairly cheap possibility, based on DragonXL:

1) Remove the DXL pressure vessel, leaving just the core Dragon bus.
2) Replace the pressure vessel with a PAF.
3) Put the module on the PAF.
4) FHE sends mutant DXL+module to a BLT-energy TLI.
5) DXL bus provides manuevering and attitude control to get the module to NRHO and berthed.
6) After berthing (or at least after grapple), the DXL is jettisoned.

A stripped-down DXL like this probably has a wet mass of 3t-4t.  FHE can put about 15t to BLT, so this should handle all of the current designs.

You could probably make a similar modification to a Cygnus.  You only need about 80m/s of delta-v.

Lots of devils in the details, obviously.  But at the end of the day, as long as the top of the stripped-down DXL can emulate the USA PAF, this seems... not exactly zero work on the part of T-A, but probably a lot cheaper than waiting to launch the stupid thing until 2028-9, or however long it takes to get Block 1B up and running.

[mordroberon]

Reviving this thread

Got into a somewhat heated discussion of reddit. Mostly constructive, but I was arguing in favor of the NRHO architecture of the mission. Here's one of the replies.

Yes, it generally takes more delta v to maintain LLO than NRHO (neither are stable), but it still isn't particularly onerous. NASA's Lunar Reconaissance Orbiter (originally intended for a 1 year primary mission) has been maintaining low lunar orbit for 15 years. There are also much more stable "frozen orbits" at certain inclinations, including at 86 deg for polar access. The detour to NRHO adds hundreds of m/s of delta v to what is required of the HLS versus doing the landing from LLO, a fraction of which could be spent station keeping in LLO. Orion's shortcoming is offloaded onto the HLS for added mission complexity and delta v.

What he says makes a good deal of sense to me. Anyone here know a good reason why NASA is going for NRHO as opposed to this lower orbit? Only reason I can think of is NRHO keeps a similar orientation wrt earth-moon. Is it just a matter of delta-v? Seems like one would be needlessly risking astronauts by stranding them on the lunar surface for a week or two before being able to rendezvous with the return craft.

[whitelancer64]

Reviving this thread

Got into a somewhat heated discussion of reddit. Mostly constructive, but I was arguing in favor of the NRHO architecture of the mission. Here's one of the replies.

Yes, it generally takes more delta v to maintain LLO than NRHO (neither are stable), but it still isn't particularly onerous. NASA's Lunar Reconaissance Orbiter (originally intended for a 1 year primary mission) has been maintaining low lunar orbit for 15 years. There are also much more stable "frozen orbits" at certain inclinations, including at 86 deg for polar access. The detour to NRHO adds hundreds of m/s of delta v to what is required of the HLS versus doing the landing from LLO, a fraction of which could be spent station keeping in LLO. Orion's shortcoming is offloaded onto the HLS for added mission complexity and delta v.

What he says makes a good deal of sense to me. Anyone here know a good reason why NASA is going for NRHO as opposed to this lower orbit? Only reason I can think of is NRHO keeps a similar orientation wrt earth-moon. Is it just a matter of delta-v? Seems like one would be needlessly risking astronauts by stranding them on the lunar surface for a week or two before being able to rendezvous with the return craft.

LRO is very close to that polar "frozen orbit" and has been fortunate to be able to be quite frugal with its fuel.

It is primarily a matter of dV. Orion's SM is not capable of getting to a ~100 km LLO and then back to Earth. IIRC the lowest circular lunar orbit it can get to while still having enough fuel to go back to Earth is about 2,000 km.

NRHO also has the benefit that Gateway / Orion / HLS are visible to Earth for communications 100% of the time.

[jongoff]

Reviving this thread

Got into a somewhat heated discussion of reddit. Mostly constructive, but I was arguing in favor of the NRHO architecture of the mission. Here's one of the replies.

Yes, it generally takes more delta v to maintain LLO than NRHO (neither are stable), but it still isn't particularly onerous. NASA's Lunar Reconaissance Orbiter (originally intended for a 1 year primary mission) has been maintaining low lunar orbit for 15 years. There are also much more stable "frozen orbits" at certain inclinations, including at 86 deg for polar access. The detour to NRHO adds hundreds of m/s of delta v to what is required of the HLS versus doing the landing from LLO, a fraction of which could be spent station keeping in LLO. Orion's shortcoming is offloaded onto the HLS for added mission complexity and delta v.

What he says makes a good deal of sense to me. Anyone here know a good reason why NASA is going for NRHO as opposed to this lower orbit? Only reason I can think of is NRHO keeps a similar orientation wrt earth-moon. Is it just a matter of delta-v? Seems like one would be needlessly risking astronauts by stranding them on the lunar surface for a week or two before being able to rendezvous with the return craft.

LRO is very close to that polar "frozen orbit" and has been fortunate to be able to be quite frugal with its fuel.

It is primarily a matter of dV. Orion's SM is not capable of getting to a ~100 km LLO and then back to Earth. IIRC the lowest circular lunar orbit it can get to while still having enough fuel to go back to Earth is about 2,000 km.

NRHO also has the benefit that Gateway / Orion / HLS are visible to Earth for communications 100% of the time.

Yeah it's primarily trying to make a location that works with SLS/Orion, because changing either of them is ridiculously expensive. The original CxP architecture from almost 20yrs ago now was going to use the LSAM to brake into LLO, so Orion was only sized with prop tanks big enough for that TEI maneuver. I agree that otherwise NRHO has a lot of drawbacks for reusable lunar landers (~5.5km/s round trip dV vs ~4km/s round trip dV for LLO). Though to be fair, if you design a system to work at NRHO, it gives you the option of boosting performance by doing a tanker refueling in LLO either on the way down, the way back, or both.

~Jon

[Emmettvonbrown]

From TLI, going IN and OUT of a low lunar orbit (100 km high)  takes 1100 m/s, twice; total 2200 m/s.
Apollo Service Module  SPS had 2500 m/s worth of delta-v (from memory) and thus could do it.
Orion SM only has 1200 m/s worth of delta-v, so can't do it: for the reasons Jon Goff explained above.
And that's the reason why they picked the NRHO in the first place, and then added the Gateway there, as some kind of intermediate step.
Note that every single Gateway iteration from 1999 to 2018 was NOT in NRHO but either at EML-1 or EML-2: much better places.
So the Gateway is also a loser in the SLS-Orion expensive boondoggle.

[sdsds]

In defense of Gateway and the 9:2 resonant NRHO: Artemis is a "Moon to Mars" initiative, and having it on a trajectory less tightly associated with lunar gravity offers 'Mars-foward' opportunities unavailable in e.g. LLO. In addition to the reasons given above, this approach somewhat appeases the 'Mars-first' crowd.

[VSECOTSPE]

Ideal staging orbits depends on your Mars transit stage propulsion crystal ball. 

If serious about electric, there’s logic to staging from an orbit that’s high in the gravity well, like NHRO (but Lagrange points are better).  But beyond the electric propulsion for Gateway itself, there’s not a whole lot of evidence that NASA is serious about electric.  It requires an order or two magnitude leap over current electric propulsion power levels.  And to my knowledge, NASA is not putting anything significant into meeting the big electric propulsion challenge.

If serious about chemical, then NHRO (or Lagrange points) are dumb as dirt and the logic is with staging from LEO or another close Earth orbit.  But that requires cryogenic propellant handling and storage at scales and on timelines that again are orders of magnitude beyond prior art.  Even an SLS-based architecture would require something around at least 10 refueling launches based on Mars DRM 5.0.  But beyond its investment in the Lunar Starship lander architecture, to my knowledge, NASA is is not putting anything substantive into meeting the big-scale cryo challenge.  (And was actually forbidden from doing so for years by Senator Backwards Shelby.)

What Mars transit stage propulsion is NASA putting semi-serious dollars into?  Nuke thermal, because  inertia, because it keeps MSFC people employed, and because of an annual Alabama earmark or two.  Probably unnecessary and egregiously expensive just like SLS if ever fully pursued.  And would still need large-scale/long-term cryo handling and storage anyway, which I guess we’re assuming can be begged/borrowed/stolen from Blue Origin if/when they get their act together.

The bottom-line is that there is no substantive NASA Moon-to-Mars plan or strategy.  A real one would be putting some serious dollars into these long tent pole technologies with clear downselects  and criteria as time went on.  Instead, we have an uncoordinated program of haphazard investments — an NHRO station here to service an underperforming capsule, big cryo work there as part of a lander program, and unrequested nuke thermal over there due to a congressional earmark.  Zero strategic thought and decisionmaking has gone into this, and to expect anyone to predict an ideal staging orbit in such an environment is folly.

And what would human Mars missions need even more than proven transit propulsion technologies?  Rapid heavy lift.  Again, even the old Mars DRM 5.0 required something approaching ten SLS-scale launches in under about 18 months.  How often is SLS projected to fly?  Once every year or three.  Again, if we were serious about Moon-to-Mars, NASA would be investing in vehicles like Starship, New Glenn, or the equivalent.  But we’re not doing that, so we’re not serious about Moon-to-Mars and this discussion is really theoretical angels-on-the-head-of-pin discussion.

Not that I’m trying to shut the discussion down, but folks should also understand the reality here that Moon-to-Mars is a slogan for a bunch of uncoordinated and haphazard spending, not an actual, thoughtful, executable engineering development program plan. 

FWIW…

[sdsds]

[...] The bottom-line is that there is no substantive NASA Moon-to-Mars plan or strategy. [...]

There's a M2M office with a M2M architecture effort that promises to deliver updates annually. That's not haphazard. Only an effort like that, tying high level objectives to implementation elements, has a chance of changing which elements are used in the long term, while still accommodating the reality of what elements are available/mandated now.

[VSECOTSPE]

There's a M2M office with a M2M architecture effort that promises to deliver updates annually. That's not haphazard. Only an effort like that, tying high level objectives to implementation elements, has a chance of changing which elements are used in the long term, while still accommodating the reality of what elements are available/mandated now.

No offense, but the “architecture” is not an architecture.  It’s just a taxonomy that sets no priorities among 60+ objectives and just logs possible elements without making any investment decisions or providing a process for how the agency is going to get to a set of priorities and choose from among potential investments.

When the M2M office can express the program’s top 3-6 objectives (not 60) and articulate the handful of key technologies, systems, or capabilities it needs to develop in order to achieve those goals, then it will have the makings of an architecture.  Until then, it’s a yearly make-work exercise that pales in comparison to annual undergraduate architecture exercises done in university aerospace engineering departments across the nation.

We did Apollo by pursuing one simple goal.  Artemis/M2M doesn’t have to be that simple.  But 60 objectives with no prioritization is unmanageable.

We did Apollo by making key decisions within a year or two on LOR/EOR, Saturn engines, and similar priorities.  Artemis/M2M doesn’t have to move that quickly.  But it needs to stop spinning its wheels on taxonomy and start winnowing, or have a plan for trade studies, demonstrations, downselects, and the like to get to those decisions in a robust and defensible way.  Without that, Artemis/M2M will remain a random assortment of investments imposed by external politics rather than an integrated architecture capable of doing something worth the cost on the Moon and getting boots on Mars before the heat death of the universe.

Again, I’m not trying to prevent discussion about Artemis/M2M alternatives here.  Lord knows the agency needs to actually examine some.  But the current NASA leadership/bureaucracy has not produced an actual M2M architecture and repeating their taxonomic gap analysis annually ad infinitum won’t produce one either.

[yg1968]

The bottom-line is that there is no substantive NASA Moon-to-Mars plan or strategy.  A real one would be putting some serious dollars into these long tent pole technologies with clear downselects  and criteria as time went on. 

The Mars part of the Moon to Mars program won't be funded for some time. There isn't enough funding for that now. The idea of looking at Mars now is only to inform what you are doing on the Moon.

[VSECOTSPE]

The Mars part of the Moon to Mars program won't be funded for some time. There isn't enough funding for that now. The idea of looking at Mars now is only to inform what you are doing on the Moon.

Sure, but Mars is not informing much of anything NASA is doing with Artemis. 

Serious consideration of options and decisions on the first big Mars tent pole — an HLV or two that can pull off a dozen or two launches within 18-24 months — isn’t being informed by Artemis.  Rather, sleepwalking on continued use of SLS in Artemis is an obstacle to serious consideration of HLV options and decisions for Mars.

And serious consideration of options and decisions on the second big Mars tent pole — a plan for investing and downselecting from among different transit propulsion options — isn’t being informed by Artemis, either.  Instead of carefully saying Artemis can do it this way or that way, but Mars needs it that way, the program is just doing a random walk gathering bits and pieces of potential solutions to the transit propulsion problem.

Even setting aside those two elephants in the room, the Deputy Administrator and the M2M office have been doing their annual taxonomy exercise for what, two or three years now?  When looking forward to Mars, has anything substantive come out of that enormous exercise that has changed what Artemis is doing?  If not, then what is the point?  Stop wasting so many careers on this Potemkin make-work exercise and just keep running the program over a cliff as a haphazard collection of politically motivated odd-ducks.

Engineers are supposed to do trades and make decisions about future capabilities and then use that to guide current investments towards worthy and executable programs.  None of that is happening here.  They’re just collecting and sorting long lists of goals and program elements.

[deltaV]

What Mars transit stage propulsion is NASA putting semi-serious dollars into?  Nuke thermal, because  inertia, because it keeps MSFC people employed, and because of an annual Alabama earmark or two.  Probably unnecessary and egregiously expensive just like SLS if ever fully pursued.  And would still need large-scale/long-term cryo handling and storage anyway, which I guess we’re assuming can be begged/borrowed/stolen from SX if Lunar Starship works.

Aside: nuclear thermal designs usually use liquid hydrogen hence the cryo handling and storage tech may be better begged/borrowed/stolen from someone with liquid hydrogen experience such as Blue Origin and its Blue Moon subcontractors or ULA, not SpaceX.