ULA claim gap reducing solution via EELV exploration master plan

[neilh]

With ISS at 2.1B per year, there still is not adequate funding for landers, outposts, and Mars, at the PRESENT time, since the vehicles need development $$ first, which is likely why the flexible path option was presented, not because it was a reason to stop HSF.

If you're correct, then one wonders why they didn't use some version of this argument in prsenting Flexible Path.  If they did, a link would be nice.

See figure 6.6.2-2 of the Augustine report (attached).

Also from the report:

Schedule also favors the Flexible Path scenario.  The
fundamental economics of the investment by NASA to
begin flights on the Flexible Path and Moon First options
are shown in Figure 6.6.2-2. Before lunar exploration can
begin, NASA must complete four development programs: 
the heavy-lift launcher, the Orion capsule, the Altair lander,
and at least some of the lunar surface systems.  Even the
well-funded Apollo Program only had to complete the first
three of these.  In contrast, exploration on the Flexible
Path could begin with just the capsule and launcher, and
then slowly develop much less costly in-space propulsion
stages and habitats.  After NASA explores on the Flexible
Path for a half decade or so, it could then invest in the lunar
lander and surface systems. In summary, the Flexible Path
provides for exploration beyond low-Earth obit several
years earlier, and allows a less demanding programmatic
investment profile.

[alexw]

You want charts?
Lets try these... 

    Thank you very much for the charts, Ross.

   But something seems wrong: for n launches/yr, the total cost of n AVH or 3n AV401 (ie, equal numbers of CCBs) is about the same. That puts the marginal cost of Centaur about zero. 

 
    Also, a question about assumptions: for the AV401 and AVH curves, do you include the fixed costs of carrying Aerojet's SRM capacity at a production rate of zero?
     -Alex

[Proponent]

You want charts?

Lets try these...  There isn't much point in keeping them close-hold any longer.

And the analysis that backs up these numbers is... where, exactly?

[Proponent]

You want charts?

Lets try these...  There isn't much point in keeping them close-hold any longer.

And the analysis that backs up these numbers is... where, exactly?

Why is it that when I ask a question about the analysis of DIRECT's costs, the answer tends to be a deafening silence?  Is it regarded as gauche to raise this subject?

[Ben the Space Brit]

You want charts?

Lets try these...  There isn't much point in keeping them close-hold any longer.

And the analysis that backs up these numbers is... where, exactly?

Why is it that when I ask a question about the analysis of DIRECT's costs, the answer tends to be a deafening silence?  Is it regarded as gauche to raise this subject?

Try sending Ross and Chuck a PM.  They are busy guys and might not see every request for information out on the public forums.  There might also be some confidential data that restricts how much of the calculations and analyses they publish in the public domain.

[neilh]

You want charts?

Lets try these...  There isn't much point in keeping them close-hold any longer.

And the analysis that backs up these numbers is... where, exactly?

Why is it that when I ask a question about the analysis of DIRECT's costs, the answer tends to be a deafening silence?  Is it regarded as gauche to raise this subject?

Try sending Ross and Chuck a PM.  They are busy guys and might not see every request for information out on the public forums.  There might also be some confidential data that restricts how much of the calculations and analyses they publish in the public domain.

Actually, the response I've seen elsewhere (I think in the DIRECT thread) is that the analysis depends on proprietary information from some of the vendors. That sounds reasonable, although it unfortunately also means that there's no way to validate the numbers or understand what assumptions went into them.

[madscientist197]

Why is it that when I ask a question about the analysis of DIRECT's costs, the answer tends to be a deafening silence?  Is it regarded as gauche to raise this subject?

And you might also ask why the remaining ESAS appendicies haven't been released. All sides seem to hide their cost data -- it is certainly not unique to DIRECT. I do not think it strengthens their argument, though.

It's very difficult to prove/disprove any cost numbers, even were all the information available. For example, an official agency could easily publish a 'cost data' document giving summary estimates of costs per year for a development programme. How would you know whether it was right or not? Sure you can check that all the totals add up, but the real devil is in the details. Do you really think they can include that information? How do you know what is a reasonable assumption or not? What percentage of the numbers have been plucked out of the air by experienced managers? (Depending on the manager, this might be either very accurate, or horrendous.) Do you expect there to be a massive library of documents proving every single assumption used in the costing?

Myself, I suspect that it's all smoke-and-mirrors (in general, not just talking about DIRECT here) and most NASA cost estimates are at best weakly based on historical precidents. In this respect I think NASA's history of credible cost-estimates speaks for itself.

[Proponent]

Try sending Ross and Chuck a PM.  They are busy guys and might not see every request for information out on the public forums.  There might also be some confidential data that restricts how much of the calculations and analyses they publish in the public domain.

Actually, I've posted similar queries in four different places recently, including the active DIRECT thread and a separate thread.

Even if there is some confidential data, how about releasing other parts of the analysis, or at least an outline?

Funny you suggest PMing.  In another thread recently, I queried claims on the part of DIRECT's supporters that DoD had multiple payloads to fly on an HLV (HLV meaning much bigger than Delta IV Heavy).  Someone suggested I PM the DIRECT people to get the explanation.  Well, I didn't want to have the kind of explanation that could only be whispered in my ear, I wanted to see some publicly-verifiable facts.  So I started a thread on the topic.  Well, that thread elicited a response repeating the claim about DoD's need for HLV but offering no evidence.  Meanwhile, one reply provided good reasons to believe that no DoD interest exists.  Yet I still see the claim about DoD's need for HLVs popping up.

Myself, I suspect that it's all smoke-and-mirrors (in general, not just talking about DIRECT here) and most NASA cost estimates are at best weakly based on historical precidents. In this respect I think NASA's history of credible cost-estimates speaks for itself.

Yeah, I guess that's kinda what I'm thinking.  It just strikes me that in the case of DIRECT, people are unusually credulous.

[Cons]

and most NASA cost estimates are at best weakly based on historical precidents.

Here is what the DIRECT team did (Jupiter-130 cost estimation):

They started with NASA's shuttle development cost estimates from 1974 (GAO 1975)
and adjusted the numbers to 2010 dollars.


                 (1971 Dollars in Millions)  (2010 Dollars in Millions)
Orbiter                   2,711          -->          15,965
3 Orbiters                1,000          -->           5,890
Systems Managment           931          -->           5,484
Contract Administration      61          -->             362   
Launch and Landing          373          -->           2,195
Configurations               90          -->             528
Main Engine                 565          -->           3,330
Solid Rocket Booster        236          -->           1,390
External Tank               183          -->           1,080
NASA Facilities             300          -->           1,767


Jupiter would reuse shuttle components so only a percentage of these
amounts would be required.


Orbiter                      5% (Avionics)               798   
3 Orbiters                  10% (Two Test Flights)       589
Systems Managment           30%                        1,645
Contract Administration     17%                           62
Launch and Landing          20%                          439
Configurations              17%                           90
Main Engine                  5%                          166
Solid Rocket Booster         5%                           70
External Tank              150%                        1,620
NASA Facilities             50%                          884
                                                       -----
                                                       6,362

And finally they added 25% margin --> $7,952 Mio.

Major concern with this approach: Aerospace inflation is
a lot higher than consumer price index inflation.

Just for comparison:
Total cost of SRB development would be $1.4 billion in 2010 dollars
according to this analysis.

But going from 4 to 5 segments alone costs more than $1.8 billion today.
http://www.nasa.gov/home/hqnews/2007/aug/HQ_C07036_Ares_first_stage.html

Predicting aerospace development costs based on consumer prices is flawed.

[Proponent]

They started with NASA's shuttle development cost estimates from 1974 (GAO 1975) and adjusted the numbers to 2010 dollars.

Thanks, that's very interesting.

Major concern with this approach: Aerospace inflation is a lot higher than consumer price index inflation.

I re-did the numbers using the NASA New Start Inflation Index (NNSII).  The NNSII's values for FY1971 and FY2010 are respectively 1.762 and 11.666, so the cost ratio is 6.62.  Applying this to the GAO's figures and DIRECT's percentages gives a total cost of $8935M, including the 25% margin.  That's not much higher than what DIRECT reports.  But, as you point out:

Just for comparison:
Total cost of SRB development would be $1.4 billion in 2010 dollars according to this analysis.

But going from 4 to 5 segments alone costs more than $1.8 billion today. http://www.nasa.gov/home/hqnews/2007/aug/HQ_C07036_Ares_first_stage.html

This exceeds even the NNSII-inflated value ($1.56G) of the GAO figure ($236M) for the cost of development of the entire SRB.  And, of course, It suggests that the NNSII's relationship to actual costs is pretty loose.

Off hand, historical prices for similar systems strike me as a reasonable starting point.  Apparently such estimates, however, need to be taken with a large grain of salt, like a factor of a few.  I also wonder why DIRECT would start from mid-1970s estimates of Shuttle development costs; surely actual historical figures are available somewhere.  And unless there's another relevant GAO study that DIRECT is referring to, I fail to understand how the 1975 GAO study is consistent with the statement:

Don't forget that GAO also says the dev costs would be about $3bn lower than we do.

Bottom line:  based on the information we have in this thread, it seems there could be a substantial risk that development of DIRECT is going to be much more expensive than many think, thereby eating the HSF budget to the point that there's little money left for payloads or missions.  It would be very helpful if DIRECT could lay out at least in outline its approach to costing.

[Warren Platts]

Yeah, and demonstrate that the marginal launch costs are going to be so much lower than the pre-existing launch fleet, that it will justify the development costs of the new rocket.

[khallow]

I re-did the numbers using the NASA New Start Inflation Index (NNSII).

I am very leery of the New Start Inflation Index for the reason that the index is used as a significant input in funding calculations for future contracts. The resulting feedback mechanism looks likely to exaggerate cost changes, both up and down. Since inflation has been the state of things since some point in the 30s, it means that there is some unknown inflation factor from this feedback effect over the lifetime of the index.

Having said that, my view is that there should be some increase over usual inflation from regulatory and liability changes since 70s and the maturation of the aero side of aerospace.

[Chris Bergin]

Follow up:

NASA and Commercial industry combine to outline FTD Propellant Depot plan:
http://www.nasaspaceflight.com/2010/08/nasa-commercial-combine-outline-ftd-propellant-depot-plan/

Discuss the above article in its specific thread:
http://forum.nasaspaceflight.com/index.php?topic=22437.0

[Proponent]

I fail to understand how the 1975 GAO study is consistent with the statement:

Don't forget that GAO also says the dev costs would be about $3bn lower than we do.

I think I can now make sense of this quote if kraisee is referring to the 2006 GAO study "Alternatives for Future U.S. Space Launch Capabilities" (attached, again).  This study quotes a figure around $8 billion for the development of a Shuttle-derived super heavy.  DIRECT, if I recall correctly, is claiming about $11 billion, a difference of $3 billion.

If I have identified the correct GAO report, I can see why the DIRECT folks would not really want to talk about it.  In both of the launch-rate scenarios considered, GAO finds that to 2017 costs for an in-line SDHLV are higher than for side-mount, despite the greater number of launches needed with the less capable side-mount system.

EDIT:  I just realized one fundamental problem with my suggestion that the report kraisee's referring to is the 2006 report "Alternatives for US Space Launch."  That problem is that "Alternatives" was written by the CBO, not the GAO.  So it would seem that either kraisee has confused the GAO and the CBO, as I did, or he's actually referring to the 1975 GAO report, which doesn't really support his claim that the GAO says DIRECT's development costs would be billions less than the DIRECT team claims.

[Proponent]

There's actually a bigger problem for DIRECT in the 2006 GAOCBO study than the comparison with side-mount, namely the comparison with Atlas V Super-heavy.  Here it is:

Vehicle

LEO Payload

Dev Cost

Unit Cost

SDHLV	125 tonnes	$8.9G	$1.3G
Atlas 5 Super-Heavy	135 tonnes	$9.0G	$1.2G

Atlas V Super-Heavy costs very slightly more to develop, but the cost difference is made up after a single launch, which carries 10 tonnes more than SDHLV.

Of course none of these costs is actually known with any precision, and the SDHLV considered by GAO is not precisely the same as what DIRECT proposes.  But it sure ain't an argument for DIRECT.

[marsavian]

I would say EELV (super heavy) has an even bigger problem, it has had no political support since OSP was canceled. Fix that and you are good to go.

[Ben the Space Brit]

I would say EELV (super heavy) has an even bigger problem, it has had no political support since OSP was canceled. Fix that and you are good to go.

And there's the rub.  Ultimately, I think that upgraded EELVs and EELV-derived HLVs, using the ACES family of upper stages and having a max-launch five-core configuration, could ultimately prove to be a lot more cost-efficient than SD-HLV for cargo launch, thanks to production line economies of scale.  However, unless some way could be found to save jobs in Utah, Texas, Louisiana and Florida (perhaps by flying crewed vehicles from a rebuilt LC-39) it is a political impossibility.

[marsavian]

George Sowers, ULA’s vice president of business development makes the case for EELV Heavy evolution against SD-HLVs.

http://www.decaturdaily.com/stories/Decatur-loses-out-in-NASA-bill,68150

[Patchouli]

There's actually a bigger problem for DIRECT in the 2006 GAO study than the comparison with side-mount, namely the comparison with Atlas V Super-heavy.  Here it is:

Vehicle

LEO Payload

Dev Cost

Unit Cost

SDHLV	125 tonnes	$8.9G	$1.3G
Atlas 5 Super-Heavy	135 tonnes	$9.0G	$1.2G

Atlas V Super-Heavy costs very slightly more to develop, but the cost difference is made up after a single launch, which carries 10 tonnes more than SDHLV.

Of course none of these costs is actually known with any precision, and the SDHLV considered by GAO is not precisely the same as what DIRECT proposes.  But it sure ain't an argument for DIRECT.

One big problem here with both is the unit cost.
What's needed is an HLV that can have a unit cots of 500M to 700M.
Otherwise it's going to be cheaper to use several 10 to 40T payload MLVs and launch a mission in pieces.

[Ben the Space Brit]

@ marsavian,

I think that Mr. Sowers is being unnecessarily pessimistic.  DoD will still want Delta-IV and NASA (and a few others) will still want Atlas-V, so neither vehicle is likely to go out of production any time soon.

Atlas-V is generally expected to be used for launching the CST-100 and Dreamchaser (in the 502 and 432 versions respectively).  That means there is work for the deployment of DEC ahead, plus possibly increased LV production for both crew launch and cargo launch for operations like Bigelow.

Looking further ahead, something similar to the ACES family of common upper stages is likely to be selected for the propellent depot/transfer tech development program.  That could mean more work, especially if ULA get the contract to build ACES; Even if they are not permitted to develop or produce the depot conversions (as these are 'spacecraft' rather than rocket stages), ULA still might buld the basic hulls for shipping to the integrator.  ACES has also got to be a front-runner for the in-space propulsion module (CPS?) contract, given that DRM 5.0 seems to have been written around something not dissimilar to the ACES-71.

Okay, so they haven't won Atlas-V Phase 2 or the 7m-diameter Super-Delta.  However, there is still a lot of projects up for grabs.  In any case, if either NASA makes a hash of SLS or the House insists on Ares-I, commercial providers like ULA win by default anyway.  It's just a matter of waiting until the current PoR crashes and burns (hopefully only figuratively).