The CBO jumps into the launcher fray

by guest blogger (and Lunar Librarian) Ken

It looks like the Congressional Budget Office decided to jump into the launch vehicle fray, releasing this week a report entitled “Alternatives for Future U.S. Space-Launch Capabilities”. (Hat tip to the ever dependable NASAWatch)

This study was prepared at the request of the “Ranking Member of the House Budget Committee” (googling…looks like that’s John Spratt of South Carolina. I have absolutely no idea if that’s of any political relevance or not). The CBO looked at six potential launch architectures, each in isolation, as if none of the other options were available to be used in a pinch (since they wouldn’t be funded by Congress). These are introduced in the summary as:

1) Pure Atlas-derived. A5M+ crew/A5H+ cargo
2) Pure Delta-derived. D4M+ crew/D4H+ cargo
3) Pure Shuttle-derived. 5Shaftd crew/Side-mount cargo
4) Pure Atlas-antecedent. A5M+ crew/ASH cargo
5) pure Delta-antecedent. D4M+ crew/DSH cargo
6) Shuttle-derived Super-Heavy. 5shaft crew/SDHLV cargo

The assumed lift to LEO for each was (shuffled by capacity):

D4M+ – 18mt
A5M+ – 24 mt
5shaft – 26 mt
D4H+ – 40mt
A5H+ – 74 mt
Side-mount – 77 mt
SDHLV – 125 mt
ASH – 135 mt
DSH – 146 mt

The report then goes into a market study in Chapter 1. The main points are:
-“Current projections indicate that maximum worldwide launch capacity for payloads less than 25mt will exceed demand by up to 100% for the foreseeable future.”
-U.S. supply and demand are assumed at 40% of global markets through 2010 [about where we are right now], then dropping to 35% because (a) our satellites will last longer [cough] and (b) Asian markets are expected to pick up.
-The only thing needing lift to LEO of more than 25mt for the next decade and a half is NASA’s implementation of the VSE.

Here we find out that the baseline mass for the NASA Lunar program is 150mt. The explanation for this seems to digest down to the fact that Apollo lifted 140mt to LEO, we’re pretty much doing the same thing, but with more crew and supplies, but lighter materials and electronics, so only a 10% increase is needed so we need, like, 154mt in LEO this time around.

The report does note that fuel represented about 75% of that 140mt, which is a key point. This also fits with a rule-of-thumb I seem to remember seeing, that it’s a roughly 5:1 mass in LEO/mass on Moon ratio (or about 80% fuel).

The usual Horowitzian objections are raised to multiple launches:
-time to execute
-leakage of cryogens during those long, lonely loitering months in orbit.
-Assembly of subcomponents in orbit is Haaard.
-So’s fully-automated rendezvous & docking.

The report also makes the very salient point that for future NASA MARS missions the mass in LEO requirements have ranged from 470-1500mt. So once NASA is done with the Moon they STILL have to face the task of on-orbit assembly to meet their “next” objective (which the report also notes was not a requirement set in the VSE – these guys really did try to lay out all the facts and be impartial). All NASA’s doing with ESAS is pushing the learning curve farther into the future. Penny wise, pound foolish is an old folk wisdom that comes to mind.

Next up is chapter 2, which fleshes out the three primary alternatives – Atlas, Delta, Shuttle. Existing capabilities are explored, and the usual crew-rating objections are raised regarding the Atlas and Delta. Here is where we find that since six launches are needed for the 150mt requirement there’s about a 10% chance of failure of any one of those missions. The implication being that if there’s mission-critical components on the launch that failed, the whole mission is a loss and the successful launches have been in vain.

It’s quickly realized that some kind of modifications will be necessary irrespective of the option chosen, so we then explore some of the componentry changes to the launchers, and what the resultant lift capability will be. The report is full of comparison tables and graphics, one of which looks at the recurring and unit costs of each option. For development, the non-recurring costs are assumed to be:

5shaft/Side-mount: 9.5Bn
D4M+/D4H+: 8.5Bn
A5M+/A5H+: 9.3Bn

For some reason the A5M+ needs 5.3Bn of mods versus 3.3Bn for the D4M+.

Unit costs for a 150mt Lunar mission (one launch of which is crew):
1 x 5shaft/2 x Sidemount: 2.35Bn per
1 x D4M+/4 x D4H+: 2.8Bn per
1 x A5M+/2 x A5H+: 1.6Bn per

The problem for Boeing is that the A5H+ has a 34Mt throw advantage over the D4H+, requiring 4 launches to get over the 150mt hurdle. Looked at programmatically, the report indicates 3 x D4H+ launches with a limit of 120mt in LEO.

It then looks at the big guys, the super heavies, the Magnum lift. Nebuchednazzer even. (sorry, that’s a champagne joke) The SDHLV at 125mt, the DSH at 146mt and the ASH weighs in at 135mt. All are very close for the unit recurring costs, at 1.3Bn, 1.5Bn and 1.2Bn respectively.

The report rolls this up into a programmatic projection out to 2020 of the costs of NASA’s implementation of the VSE, on both ambitious and less so bases. They all come out about the same, although there’s a lot that can be read into what’s presented.

Box 2.1 notably goes into the risks associated with developing and producing the NASA choices (since they’ve been harping on the Delta and Atlas shortcomings so far). These are:

-Stability control
-Large second stage powered by modified engine
-Structural integrity of 5shaft
-Organizing and executing the construction of a launch vehicle rated for manned space flight [!?!]

Appendix A looks at on-orbit assembly. It reiterates that the US has not demonstrated autonomous rendezvous & docking, and the more launches one has the greater the risk of something happening to one of them that threatens the mission.

But let’s step back for a minute and remember what the report said earlier about the composition of the launches – about 75% fuel. Of course, the A5H combo seems to be on par with the NASA proposal as far as number of launches goes, so the main risk seems to lie with Boeing, which has 5 launches required. But three of those are obviously fuel, unless you’re stupid enough to ship the computer boards with the fuel supply. Looking at the launch manifests again, we see:
Shuttle: 1 crew, 1 cargo & fuel, 1 fuel
Delta: 1 crew, 1 cargo & fuel, 3 fuel
Atlas: 1 crew, 1 cargo & fuel, 1 fuel

So ultimately your core risks are the same: 1 crew launch and one cargo launch that you really, really have to worry about, and then the fuel launches, which shouldn’t fail but if they do…ehh. File an insurance claim and roll up the next one.

Appendix B looks at Human Safety factors. It gives us some NASA Standard acceleration limits, notes that modified EELVs likely wouldn’t exceed those, and notes that Apollo 17 would have (slightly).

Appendix C lists the Considerations affecting multiple launch Lunar missions:
-Launch Cycle
-Launch Delays
-Commercial access to Delta and Atlas launch capability

Appendix D is a bunch of pretty powerpoint-ready pictures of how the vehicles evolve graphically, while Appendix E delves further into the cost estimates for the ambitious and less so approaches.

It’s a pithy report and well worth the read. It does provide a balanced presentation of the options, but in an unfortunately simplified way by virtue of the “pure-” paths of the options.

When Jon and I talk about launch architectures, we talk about flexibility. Make your CEV so that it can fit on the Delta AND Atlas. Provide orbital infrastructure for Moon missions AND other things of interest. Unfortunately, that kind of chaotic complexity doesn’t boil down easily into this kind of modeling, and even parametric modeling attempts at such a system would be difficult. The CBO report doesn’t consider such things as launch demand from Bigelow facilities. It only looks at how NASA is implementing their vision of the Vision, and considers that Congress will only fund one launch vehicle.

Put in that light the Shuttle-derived stuff doesn’t come out looking too bad, making the politically-easy-thing-to-do be support of the Shuttle jobs. Of course, a Statesman would realize that the business of America is business, and the commercial providers can take us to where we need to go, because they’re going there as well.

There are certain undercurrents in the report, and Boeing seems to come off the worse for it. As an analyst, I have to read the report in the context of it having been written by a political entity. In that regard they certainly tried to be fair and balanced in what was presented, but I still feel there was some bias, and sense a bit of tension in the wording. I also have to assume that since it’s related to a political entity that “there are larger forces at work here than you know”. In that context I get the feeling that Boeing is being gently sidled out of the civil human spaceflight field, putting LockMart at the fore, with Boeing probably picking up defense work in return.

From a commercial standpoint, it doesn’t convince me of the need for ESAS. The report clearly notes that existing manufacturing capabilities at the heavy end exceed market demands, now and until 2020. Why then would we want to develop a whole new heavier lift capability? It just doesn’t follow. Also, the issue with on-orbit assembly really bugs me. The NASA plan just pushes it out into the future. We need to learn this to be commercially competitive in the future. The U.S. needs to develop it so that we can sell the products and services for a fair price to everyone else. It makes sense to deal with the issue now so that WE are the skilled ones at doing it. Starting small, with depots, free-flyers, and facilities in LEO, extending out to EML-1, extending further to the Moon, further still to NEOs, furthers yet to Mars & Beyond (Ceres here I come!). Building it piece by piece, but lots of pieces starting NOW, and often, and involving both business and government.

Conclusion: There are a lot of interesting facts, but also some limiting assumptions. Any number of conlusions can be drawn from it, as it makes no attempt to draw conclusions itself. The results are ambiguous, with a slight advantage for Lockmart, a slight disfavor to Boeing, leaving the Shuttle-derived as a nice, safe, comfortable political alternative.

NB: The views, opinions, and conclusions above are solely the product of the author, and do not necessarily (and probably don’t) reflect anyone else’s. Caveat emptor, you get what you pay for.

The following two tabs change content below.

Latest posts by lunadyne (see all)

This entry was posted in Uncategorized. Bookmark the permalink.

5 Responses to The CBO jumps into the launcher fray

  1. Habitat Hermit says:

    Direct link to the report (PDF warning – 1.3 MB) here

  2. mz says:

    Of the EDS and LSAM descent stage propellants which are a majority of the stack mass, there’s liquid oxygen in about 6:1 relation…
    Meaning you just need to keep LOX in the depot, and can launch hydrogen with the respective payloads, meaning you don’t have to store hydrogen in orbit for long. LOX is easier.

  3. Jon Goff says:

    What? But if you launched the LOX first, and only launched the Hydrogen last, the boiloff issue (and hence the length of time over which you could assemble your stack) would pretty much go away! How then could they justify their PorkLauncher V program? Pay no attention to the rather pinkish color of the Emperor’s new clothes….


  4. Grant Bonin says:

    Excellent overview, Ken.

    Couldn’t agree more (as some know) with your argument that components should be designed to interface with *more than one* LV. Also, the desire to build bigger and bigger boosters to prevent your mission from out-growing your launch capability seems to be the opposite of what would be prudent–to wit, developing intuition for operations in space.

    I think my biggest issue is with many people’s assessment of the propellant boiloff issue as a showstopper. Yes, boiloff is a real consideration; but using MLI, it’s been *demonstrated* that you can restrict your rate of mass loss in an H2/LOX stage to approximately 1% per month. And if that’s not good enough, google “hybrid propellant module” or “zero boiloff” and discover that NASA itself was able to produce a zero-boiloff propellant module *a decade ago*!

    Emperor’s new clothes indeed…

  5. Will McLean says:

    Some observations:

    2020 is the wrong timeframe. You really should look at discounted life cycle cost for at least the first ten years of lunar missions. Some of the cheaper plans are no longer cheaper on that basis.

    The economics are oversimplified. Some of the versions with smaller launchers would benefit from lower unit manufacturing costs at higher flight rates. On the other hand, the increased costs after launch, for development and recurring costs of tugs and the higher costs of designing the EDS and LSAM to break down into smaller payloads are not considered.

Leave a Reply

Your email address will not be published. Required fields are marked *