Creating capability foundations for NASA’s exploration roadmap

[Steven Pietrobon]

I really want to like the concept of a 6 x RS-25 core that meets the Congressional mandate without requiring advanced boosters. But I wonder:  could the tanks sized for that core could be built on the equipment created for the 4 x RS-25 core?

I don't see why not. The tank volumes are exactly the same. An additional LOX down pipe will need to be added and a new thrust structure built. The wall thicknesses may also change, but that's just a matter of reprogramming the milling the machine.

[MATTBLAK]

Adding another 2x RS-25 engines will just take too many costly redesigns to incorporate them. The money would be better spent on better strap-on boosters - preferably liquid - and better upper stages. The RS-25E might also be a candidate for thrust uprating to the neighborhood of 115% percent or more over the traditional Shuttle engines, with little or no impact to reliability or specific impulse.

[sdsds]

The money would be better spent on [...] better upper stages.

Here we are in complete agreement!

Adding another 2x RS-25 engines will just take too many costly redesigns to incorporate them. [...]
The money would be better spent on better strap-on boosters - preferably liquid

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

[newpylong]

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

Skip all of this and just replace the steel boosters with advanced composites with newer propellant. They will have more thrust, be lighter, and cheaper to produce.

Save your money and don't touch the core except for what might be needed to support the new boosters.

[Space Ghost 1962]

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

Skip all of this and just replace the steel boosters with advanced composites with newer propellant. They will have more thrust, be lighter, and cheaper to produce.

Save your money and don't touch the core except for what might be needed to support the new boosters.

You're not saving money. Core costs - design/fixed/time/other - are much higher.  Touching the core in any way is more expensive than the boosters.

Cheaper to handle all changes in the boosters. Live by the core definitions of the boosters, die by the core definitions of the boosters.

[MATTBLAK]

The money would be better spent on [...] better upper stages.

Here we are in complete agreement!

Adding another 2x RS-25 engines will just take too many costly redesigns to incorporate them. [...]
The money would be better spent on better strap-on boosters - preferably liquid

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

"show your work" - I'm not an engineer working in the field of space exploration!! But similar opinions that I'm probably just regurgitating have occurred from much smarter folk than I in other, similar threads.

[Chris Bergin]

better upper stages.

That is a box about to be ticked off. Should be our next SLS article, via Chris G.

It's interesting to note, that by now (in fact, way before this point) CxP was already reducing capability and struggling to make a square peg fit into a round shaped hole. It feels like the opposite with SLS, in fact, it is the opposite. Part of the reason I get annoyed with lazy "it's just Constellation again" comments.

Sure, anyone can twist that comment into "but it is, because..." However, via covering this vehicle - daily, via talking to the troops and reading all the documentation (of which there is more than all the documentation ever published on any commercial vehicle combined) - from an engineering, development and performance (per part of this conversation of late) standpoint, one has to feel optimistic.

[sdsds]

just replace the steel boosters with advanced composites with newer propellant. They will have more thrust, be lighter, and cheaper to produce.

Well I agree: that would be great vehicle and I can't imagine NASA really needing more than that. But it's performance wouldn't meet the Congressional requirement.

121.5 t to LEO with a a four engine core and ATK Advanced Boosters.

You only get a vehicle delivering > 130 t to orbit using ATK advanced boosters when you also add another engine to the core. From the perspective of non-recurring costs it matters little whether you add one engine or two to the core. But if you add two, then you don't need the advanced boosters.

Core costs - design/fixed/time/other - are much higher.  Touching the core in any way is more expensive than the boosters.

Cheaper to handle all changes in the boosters.

Do you mean with a liquid booster design?

You could convince me, if you show your work. Anything will do, really! Back of the envelope; whatever. Just show how it could plausibly be cheaper to develop a new engine, a manufacturing capability for it, a new tank, and manufacturing capability for that too, compared with a modest redesign of something that can be built with the extant manufacturing infrastructure. Are you assuming a cost model for the boosters that is radically different than your cost model for the core?

[TomH]

You only get a vehicle delivering > 130 t to orbit using ATK advanced boosters when you also add another engine to the core.

I could be wrong, but I thought ATK said the LV with Dark Knights would need 5 core engines and the full sized LUS with 2 x J2-X to reach the mandate. The extra mass atop the core would require a core redesign.

The use of either five or six mains on the core will burn through the propellant well before near orbital velocity is reached. This means J-2X with its higher thrust would be required to fight gravity losses on the way to parking orbit. J-2X would be restarted for earth departure burn and at this point its lower ISP becomes a disadvantage.

Four engines allow the core to burn all the way to disposal orbit. The burn of the sustainers is maximized. The remaining US burns are for circularization and earth departure. Not having high gravity losses to fight, the higher ISP RL-10 or MB-60 can be employed instead of J-2X. To increase the payload, modifications can be restricted to the boosters. Advanced solids increase that some without flame trench or LM modifications. Advanced liquids increase the payload much more, but said trench mods must be made as well as propellant umbilicals added to the ML and the non-functioning Saturn RP-1 fueling system replaced . None of the advanced boosters requires changes to the core or the EUS, though I presume the flight guidance system will need at least software modifications, if not more.

[newpylong]

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

Skip all of this and just replace the steel boosters with advanced composites with newer propellant. They will have more thrust, be lighter, and cheaper to produce.

Save your money and don't touch the core except for what might be needed to support the new boosters.

You're not saving money. Core costs - design/fixed/time/other - are much higher.  Touching the core in any way is more expensive than the boosters.

Cheaper to handle all changes in the boosters. Live by the core definitions of the boosters, die by the core definitions of the boosters.

Did you not read what I wrote? I just said this... 

[Coastal Ron]

...However, via covering this vehicle - daily, via talking to the troops and reading all the documentation (of which there is more than all the documentation ever published on any commercial vehicle combined) - from an engineering, development and performance (per part of this conversation of late) standpoint, one has to feel optimistic.

Haven't seen if you covered the most recent GAO annual assessment of “large-scale” NASA projects (link here), but there was a chart that I thought was interesting on Page 10.  My summary of it is:

When Bolden took over from Michael Griffin in 2009 the average launch delay was 11 months, and the average cost growth excluding the JWST was 12%.

Five years later the average launch delay is 3 months, and the average cost growth excluding the JWST is 3%.

I attribute this directly to the overall job Administrator Bolden is doing, and I see it affecting the SLS program too - obviously in a good way.  And that's good not only for SLS supporters, but for all of us that are U.S. Taxpayers and want to make sure our money is being spent wisely.

That said you can have the best managed and well engineered product in the world, but if no one needs it then it doesn't matter.  And the "need" part for the SLS is awaiting our politicians to validate, with no known ETA...

[JohnFornaro]

...However, via covering this vehicle - daily, via talking to the troops and reading all the documentation (of which there is more than all the documentation ever published on any commercial vehicle combined) - from an engineering, development and performance (per part of this conversation of late) standpoint, one has to feel optimistic.

...When Bolden took over from Michael Griffin in 2009 the average launch delay was 11 months, and the average cost growth excluding the JWST was 12%.

Five years later the average launch delay is 3 months, and the average cost growth excluding the JWST is 3%.

I attribute this directly to the overall job Administrator Bolden is doing...

Which is a fascinating statistical tidbit.  As Chris mentions, SLS appears to be making its milestones appropriately.

When they check off the upperstage box, that will be great too.

As always, there is nothing pragmatic on top of the rocket and no good strategy for HSF.  I don't know if progress is being made or not.

[Space Ghost 1962]

I think you could convince me if you could show your work supporting these conclusions.

What I see is that for a liquid booster, your tasks are to: develop a new engine, develop new tanks, create new engine manufacturing capability, and create new tank manufacturing capability. In contrast for a core upgrade you need to: develop a modified version of tanks you already use, and build those tanks with your existing manufacturing capability.

Comparing the costs associated with each of these two sets of tasks, how do the sums works out in your mind?

Skip all of this and just replace the steel boosters with advanced composites with newer propellant. They will have more thrust, be lighter, and cheaper to produce.

Save your money and don't touch the core except for what might be needed to support the new boosters.

You're not saving money. Core costs - design/fixed/time/other - are much higher.  Touching the core in any way is more expensive than the boosters.

Cheaper to handle all changes in the boosters. Live by the core definitions of the boosters, die by the core definitions of the boosters.

Did you not read what I wrote? I just said this... 

Note the bold above. The "except". I read it carefully. I'm saying no exceptions, period. Neither for LRBs either.

That word alone is a billion dollar plus difference.

Core costs - design/fixed/time/other - are much higher.  Touching the core in any way is more expensive than the boosters.

Cheaper to handle all changes in the boosters.

Do you mean with a liquid booster design?

You could convince me, if you show your work. Anything will do, really! Back of the envelope; whatever. Just show how it could plausibly be cheaper to develop a new engine, a manufacturing capability for it, a new tank, and manufacturing capability for that too, compared with a modest redesign of something that can be built with the extant manufacturing infrastructure. Are you assuming a cost model for the boosters that is radically different than your cost model for the core?

If you read the Advanced Boosters RFP, you'll note that the requirement is that the core is unmodified.  This is well documented as to why. One can also track the costs of successive revisions of the core and the consumed budget over each, and one can also track the collateral cost of changes to MPS, tanks, etc, the costs add up rather quickly.

So the bottom line is that the core is defined to the RSRMV. The three competitors design to this.

Now, if this were to be relaxed, the following would have to happen given contract law and related federal regulations:
A. Each could request different, optimized attachments of indefinite scope.
B. None of the three need be the same. I'd be surprised if they were even close. Even the two LRB's would be different. And I'm very sure even the SRB one would significantly differ than the RSRMV one so as to use the composite casing better.
C. To evaluate winners, three different core designs with appropriate boosters would need to be evaluated.
D. Challenges would be made by losers over the nature of the changes in the core superficially or non-effectively appearing to change performance.

Its tough enough with a  single core as it is. The more ambiguity added, complexity shoots up. The core is already too complex.

And that is why the RFP is that way. Indirectly, its also why the number of second stage options needs to drop too. More is not always better.

The only way this could have been better is if the competition happened before the core was defined, with an abstract core structure. All the four would have a different result, but the net effect could be judged apart from actual implementation of the core, so we'd get the best of all, and the ability for the core to be designed best in the context of the chosen boosters.

But this was not done.

[sdsds]

If you read the Advanced Boosters RFP, you'll note that the requirement is that the core is unmodified.

Thank you for your well-reasoned reply! FWIW I largely agree with what you wrote.

the bottom line is that the core is defined to the RSRMV. The three competitors design to this.

Yes, this is clearly best, if there is to be an advanced booster competition at all.

Now, if this were to be relaxed [...]

I elide the (very good) description following this bit, because "relaxing" the advanced booster competition is not (I hope) being discussed. Canceling the advanced boosters entirely is the suggestion at hand!

The core is already too complex.

I would like to understand this assertion better. The core is clearly more complex than an ET. But it is less complex than an orbiter. Can you expand on the aspects of the current core that make it too complex?

[Space Ghost 1962]

Now, if this were to be relaxed [...]

1. I elide the (very good) description following this bit, because "relaxing" the advanced booster competition is not (I hope) being discussed.

 Canceling the advanced boosters entirely is the suggestion at hand!

The core is already too complex.

2. I would like to understand this assertion better. The core is clearly more complex than an ET. But it is less complex than an orbiter. Can you expand on the aspects of the current core that make it too complex?

1. What I wrote was encompassing what happens in a program of this size. It all fits together. And yes sometimes parts of a competition can be relaxed, it has happened before with some of those involved. All it takes is someone from congress beating the drum for an anxious competitor - they forget that when they do this, it skewers them because the others get to play more too.

2. There's too much here for a post to be comprehensive. Let me attempt an analogy with a story first.

STS had many (hundreds, thousands) of back of the envelope designs/concepts/etc. With them a lot of issues came up, and long before actual designs came into existence.  So rather than force top down things,  there was a rich environment where any different ways of combining things made many different ways, were shared by many designers. They got used to borrowing  parts of concepts/implementation to iterate next ones better.

The problem with the core is that it isn't being done through a competition. Or iteration. Its a refined single source design, and typically these aren't brilliant just "good enough" - they serve a purpose to get a job done. There's no competitive motivation of multiple design groups to specialize in ways that reduce / improve the core. Such designs leave in too much because its safer as a designer to have them than to leave them out and tweak other things - like asymmetrical thrust structures that are arranged in a certain symmetry to achieve the same effect for example.

Wouldn't be so bad, except that dependencies multiply this as you complete all of the elements of the design. In the case of a heavy lift vehicle, there's not many to draw on, so you tend to draw off the few, older designs that don't contain much of new ideas, and there's been a lot of change in mechanical design since the last HLV. STS ET was a drop tank not a stage, and went through considerable refinement over 100+ flights, but this isn't necessarily applicable to an inline stage.

So its more complicated than it needs to be to begin with. As you refine/test/simulate/wash/rinse/repeat, you find issues and again its easier to add than to delete. Sometimes this transfers the problem to someone else's responsibility, but the addition remains. This is how we get weight/complexity growth, and typically it mounts up until it gets silly, and then whole portions are replaced to buy back some. These get increasingly harder as more and more is locked down in the core, which is good because you get somewhere, but bad because of the straitjacket for the rest. By this point FEM, aeroloads, thermal/stress analysis, results from subscale models, etc start nailing things down. Changing before actual flight data becomes precarious.

You've now got a 100 million piece jigsaw puzzle. Lets add second stage, payloads, and boosters to the "all up" vehicle in use. Now you have to track down load paths, dynamic behavior, and ... it gets very slow and messy. Remember this is in the minds/computers of hundreds/thousands that need to be coordinated. Its like walking through molasses. Very, very expensive.

As I hinted, I think second stage options will drop because of this - due to the rocket equation, they are much more difficult/delicate than the first stage by far. I can't see how you get to 130T though without considerable performance increases either in second stage or boosters.

Boosters are like simpler core stages, so the cost growth isn't like the second stage at all. In fact, if you have enough control authority in the core stage, they can be the cheapest part of the entire launch vehicle - no gimbals and limited torques/bending. Since you've already paid for the core, it becomes the "middle stage" that allows you to keep from changing the second stage, and all you do is offset the needed performance from additional booster sizing.

Either thrust or thrust+Isp - the two LRBs. The advanced solids optimize the old solids in marginal improvements of weight, propellant, grain, burn profile and duration - attempting "good enough" at the top end of what can be done.

There's a lot more to this. And as usual I've been too wordy.

PS. I've lately been thinking back on STS, going through some of the competitive ideas that were dismissed, I think that it was too rushed - there were better solutions in certain cases to what was used that there wasn't time for. The more I look at them and later application to other uses, I'm very certain we could have done significantly better within a marginal increase in time/cost. Not second guessing some people who made hard decisions - don't think things made it to that level of awareness.

Point here is that concluding how to design a complex vehicle brings its own costs/risks when you don't have broad enough debate on how to do it.

[Proponent]

NASA launch paradigm reorientation - $3-11B (no, I'm not kidding; it's from Augustine)

Specifically, what Augustine said, in 2009, was:

The EELV-heritage super heavy would represent a new way of doing business for NASA, which would have the benefit of potentially lowering development and operational costs. The Committee used the EELV-heritage super-heavy vehicle to investigate the possibility of an essentially commercial acquisition of the required heavy-launch capability by a small NASA organization similar to a system program office in the Department of Defense. It would eliminate somewhat the historic carrying cost of many Apollo- and Shuttle-era facilities and systems. This creates the possibility of substantially reduced operating costs, which may ultimately allow NASA to escape its conundrum of not having sufficient resources to both operate existing systems and build a new one. However, this efficiency of operations would require significant near-term realignment of NASA. Substantial reductions in workforce, facilities closures, and mothballing would be required. When the Committee asked NASA to assess the cost of this process, the estimates ranged from $3 billion to $11 billion over five years.

(emphasis added).  Yet, according to the GAO,

After the President proposed canceling the Constellation program in his fiscal year 2011 budget request, NASA reported that the agency's costs associated with terminating the various Constellation program contracts could reach close to $1 billion.

(again, emphasis added).  This supports darkbluenine's assertion that the $3-11 billion claim was exaggerated.

After seeing the recent GAO report on ML-2, I can't help but think that paying even $11 billion in termination costs circa 2011 would have been a bargain. The costs and schedules discussed ten years ago even by SLS's detractors were very optimistic.

[Proponent]

Specifically, what Augustine said, in 2009, was:

The EELV-heritage super heavy would represent a new way of doing business for NASA, which would have the benefit of potentially lowering development and operational costs. The Committee used the EELV-heritage super-heavy vehicle to investigate the possibility of an essentially commercial acquisition of the required heavy-launch capability by a small NASA organization similar to a system program office in the Department of Defense. It would eliminate somewhat the historic carrying cost of many Apollo- and Shuttle-era facilities and systems. This creates the possibility of substantially reduced operating costs, which may ultimately allow NASA to escape its conundrum of not having sufficient resources to both operate existing systems and build a new one. However, this efficiency of operations would require significant near-term realignment of NASA. Substantial reductions in workforce, facilities closures, and mothballing would be required. When the Committee asked NASA to assess the cost of this process, the estimates ranged from $3 billion to $11 billion over five years.

(emphasis added).  Yet, according to the GAO,

After the President proposed canceling the Constellation program in his fiscal year 2011 budget request, NASA reported that the agency's costs associated with terminating the various Constellation program contracts could reach close to $1 billion.

(again, emphasis added).  This supports darkbluenine's assertion that the $3-11 billion claim was exaggerated.

In another thread, VSECOTSPE has just provided chapter and verse for the fact that the cost of terminating Constellation contracts would have been approximately $1 billion, not the $3-11 billion suggested by Augustine.

The Augustine report's conclusion that NASA would accomplish nothing in human spaceflight beyond earth orbit without at least an additional $3 billion per year has been validated, even if Congress willingly ignored this fact for years. But, in hindsight, I can't help but think on the whole Augustine helped create the Shuttle-derived black hole NASA finds itself in today. For one thing, the report buried the lede: its gingerly-expressed statement that NASA needed to get off of Shuttle-derived launch vehicles and onto commercial rockets was deep inside the text, not in the headline. And Augustine himself, speaking after the report's release, said "Don't skimp on the heavy lift"; since the definition of heavy-lift is vague (few are aware that the report itself defined 50 tonnes to LEO as "heavy"), this statement gave self-interested congresspeople cover to advocate for SLS.

P.S. Note that the GAO report is dated July 2011, months before SLS was approved.

[VSECOTSPE]

Specifically, what Augustine said, in 2009, was:

The EELV-heritage super heavy would represent a new way of doing business for NASA, which would have the benefit of potentially lowering development and operational costs. The Committee used the EELV-heritage super-heavy vehicle to investigate the possibility of an essentially commercial acquisition of the required heavy-launch capability by a small NASA organization similar to a system program office in the Department of Defense. It would eliminate somewhat the historic carrying cost of many Apollo- and Shuttle-era facilities and systems. This creates the possibility of substantially reduced operating costs, which may ultimately allow NASA to escape its conundrum of not having sufficient resources to both operate existing systems and build a new one. However, this efficiency of operations would require significant near-term realignment of NASA. Substantial reductions in workforce, facilities closures, and mothballing would be required. When the Committee asked NASA to assess the cost of this process, the estimates ranged from $3 billion to $11 billion over five years.

(emphasis added).  Yet, according to the GAO,

After the President proposed canceling the Constellation program in his fiscal year 2011 budget request, NASA reported that the agency's costs associated with terminating the various Constellation program contracts could reach close to $1 billion.

(again, emphasis added).  This supports darkbluenine's assertion that the $3-11 billion claim was exaggerated.

In another thread, VSECOTSPE has just provided chapter and verse for the fact that the cost of terminating Constellation contracts would have been approximately $1 billion, not the $3-11 billion suggested by Augustine.

A couple things:

I’m darkbluenine.  That’s my old screen name here.  I lost the password after a long period of inactivity, and as a result, it hasn’t been used in years (and won’t be).  Bergin was kind enough to let me create a new one.

There’s some apples and oranges in these numbers.  GAO is talking only about contract termination costs, while Augustine is talking about the broader category of program termination costs, which can also include things like workforce transition, facility closure, and environmental remediation costs.

That said, I stand by my earlier post that the ginormous error bar on the $3B to $11B cost for Constellation program termination that NASA gave Augustine was ridiculous.  It either meant NASA had no handle on Constellation costs or that NASA was trying to put a thumb on the scale of Shuttle-derived options by making their termination so potentially expensive.  Probably some of both.

The Augustine report's conclusion that NASA would accomplish nothing in human spaceflight beyond earth orbit without at least an additional $3 billion per year has been validated, even if Congress willingly ignored this fact for years. But, in hindsight, I can't help but think on the whole Augustine helped create the Shuttle-derived black hole NASA finds itself in today. For one thing, the report buried the lede: its gingerly-expressed statement that NASA needed to get off of Shuttle-derived launch vehicles and onto commercial rockets was deep inside the text, not in the headline. And Augustine himself, speaking after the report's release, said "Don't skimp on the heavy lift"; since the definition of heavy-lift is vague (few are aware that the report itself defined 50 tonnes to LEO as "heavy"), this statement gave self-interested congresspeople cover to advocate for SLS.

I don’t think anyone on Augustine II was intentionally malicious in any way.  But their report certainly succumbed to too-many-cooks-spoil-the-soup syndrome.  It lacked clarity, confidence, and strong findings and recommendations.  Instead of pointing towards a clear path forward, almost anyone’s architecture or vehicle could find something to support their case in that report.  This is typical of blue ribbon committees.  If the White House cares about an issue, factions in the Administration will fight it out internally, potentially resulting in a decision paper and meeting going all the way up to the President.  But if an issue is a low priority, like civil human space exploration has been for decades, then they’ll farm it out to a balanced committee of experts with different/opposing viewpoints, which is a sure-fire recipe for getting lowest-common denominator advice.  But that’s okay, since the Administration didn’t really care about the outcome on the issue in question in the first place.

That said, when NASA handed Augustine II a cost estimate for Constellation termination with a 300%+ error bar, someone on the committee should have called “bullcrap” and told NASA to go back to the drawing board.  The fact something like that didn’t happen is an indicator of poor due diligence on the part of Augustine II.

P.S. Note that the GAO report is dated July 2011, months before SLS was approved.

This thread is quite a ride down memory lane.  Most of it has not held up well as Orion/SLS have grossly underperformed the expectations of their old advocates here.  But at the risk of quoting myself and an entrepreneur I admire, here’s a couple nuggets from 2014 in this thread that have withstood the test of the ensuing decade:

Gerst told ASAP that SLS must launch every year for flight safety.  Unless Gerst is wrong, a requirement of one launch every two years does not support a safe system.  The NAC Chair has also hit on this issue in congressional testimony.

Spreading this launch rate over an annual program runout cost of ~$1.5B, the launch cost alone of each Tactical Mission will be ~$3B.  Add in an MPCV, ground support, and mission-specific costs, and total cost for each Tactical Mission will be pushing/exceeding $5B.  Just for one 1960s-era mission around the Moon.  It's hard to see that being sustainable for more than a mission or two.

I think it isn't unrealistic to think that by the time SLS/Orion are capable of doing their first mission, that new technologies will be "on the shelf" that will make those two vehicles look as anachronistic as an 8-track player.

FWIW...