[FoRK] The Bush Space Initiative: Fiscal Nightmare or... Fiscal
Nightmare? (Space Daily)
Joseph S. Barrera III
joe at barrera.org
Thu Mar 18 01:19:09 PST 2004
Pretty depressing reading. Maybe we should just keep sending rovers to Mars.
The Bush Space Initiative: Fiscal Nightmare or... Fiscal Nightmare?
by Jeffrey F. Bell
Honolulu - Mar 17, 2004
The new space initiative announced by President Bush has the odd
distinction of being criticized both for costing too much and costing
too little. Many commentators have denounced Plan Bush an insanely
grandiose program that will waste $1 trillion dollars of tax money.
At the same time, another group of critics says that it is a trivially
small program whose funding level is utterly inadequate to achieve its
announced goals of manned flights to the Moon and Mars. John Pike goes
so far as to call the Bush Plan "a roadmap for the quiet and orderly
phase-out of manned space flight."
Surely, at least one of these schools of criticism must be based on
politics, or the commentator's personal agendas, or simple stupidity.
They can't possibly both be right -- can they?
Actually, they can. The blind men are feeling different parts of the
elephant, or more accurately the same elephant at different times in the
future. John Pike is looking at the funding plan over the next 5 years.
The budget chart released by NASA shows that the continuation of the
Station and Shuttle programs through 2010 will continue to suck up most
of NASA's budget. Only after the termination of Shuttle is there any
significant funding available for Plan Bush.
In fact, of the scant $12B supposedly allocated for the new initiative
in 2005-2010, detailed budget plans recently released by NASA show that
only $6.6B is actually for Project Constellation, the new name for the
development of the Crew Exploration Vehicle that will eventually carry
astronauts back to the Moon.
If that sounds like a lot of money to you, consider this: Boeing is
currently planning to spend about $7.5B dollars on development of its
new 7E7 airliner. The 7E7 will be a progressive development in a long
series of subsonic transports going back to the 707 of the 1950s. Boeing
and its subcontractors have been continuously improving that basic
concept in small increments for more than 40 years. There are no big
technical jumps, just an incremental improvement in efficiency and safety.
The US Army has just cancelled its RAH-66 2-seat scout helicopter after
22 years of development. This project absorbed about $8B without
producing a single production aircraft. It is estimated that "closeout
costs" might amount to another $3B. The development schedules and costs
for other military aircraft programs like V-22 and F-22 are so
depressing I can't bear to mention them here.
Project Constellation is far more ambitious. There has been essentially
no work on manned ballistic re-entry vehicles in the US since Apollo,
due to NASA's persistent attachment to winged or lifting-body designs.
The only relevant reservoir of recent design experience is with small
Rvs for planetary probes like Stardust. It's no wonder that estimates
for the total development cost of CEV hover around $15B.
But won't we be getting a much better spacecraft than Apollo? Haven't we
made all kinds of technical advances since 1963? Can't we pick a new
shape that will be kinder and gentler? Won't advanced composite
materials and electronics make the CEV much lighter than Apollo?
Well, no. The original Apollo team had the benefit of actual tests of
hundreds of possible reentry shapes made by the US and UK missile
programs in the mid-50s. They explored every possibility with
wind-tunnel and arcjet tests, and all the promising shapes got actual
flight tests at Wallops Island or Woomera.
Even the Soyuz "headlight RV" design seems to be based on one of the
rejected Apollo proposals. There is no possibility that a significantly
better shape could be found today. The only reason to develop a
different "mold line" for CEV is public relations.
And as for weight -- we can't calculate that in an absolute sense, since
the number of astronauts to be carried in the CEV is still undefined,
and won't be until a "Level 1 requirements document" is produced many
months from now. But one can expect that it will be heavier than Apollo
per capita, not lighter. Its design team will face many requirements
that Apollo's didn't:
- - Unlike Apollo, the CEV must be designed from the start to allow for
its future use on long Moon missions, and eventually as the Earth Entry
Module for Mars missions. Many weight-saving tricks used on Apollo (e.g.
the 5psi oxygen atmosphere) were only possible because of its limited
flight duration. And safety factors were cut to levels that would be
* If the CEV is to be parked for long periods in space or on the
Moon, its avionics need to withstand an intense solar flare like the one
last November that knocked out three modern unmanned spacecraft and
damaged many others. Modern microelectronics is much more vulnerable to
solar particle damage than those old Apollo transistors. Heavier
rad-hardened computers must be developed.
* For the same reason, CEV will probably use hydrogen peroxide
thruster fuel instead of the more toxic and corrosive hypergolic fuels
MMH/NTO. Experience on both Skylab and Shuttle has shown that hypergolic
thrusters begin to leak and/or fail after a week of active use, or a few
months of idle storage. Peroxide monopropellant thrusters are much less
efficient than bipropellant systems, so fuel weight will be much higher.
* We also can't use fuel cells in the CEV Service Module because of
the same space-storage requirement. All CEV concepts show big
Soyuz-style solar panels and batteries for dark periods.
* Those super-light advanced materials wrecked the X-33 program
because they were developed for airplanes, not spacecraft. They are also
so expensive that they are used mostly in super-expensive military
planes. As I will show below, CEV needs to be cheaper than Apollo.
So the fuzzy picture of CEV that evolves from this analysis is of an
Americanized Soyuz: rugged, simple, reliable, low-tech but heavy. (Maybe
we could cut out the middlemen and give the CEV contract directly to
Energia? Sorry, that's illegal.)
Another big difference between Apollo and Constellation is the funding
profile. Apollo was front-loaded with most of its funding concentrated
in the early years. While a lot of this was sucked up by construction of
facilities, there was still enough to hire an army of engineers and do
the detailed design phase on a rush basis. Constellation is on a strict
diet until at least 2010. This is what John Pike means when he says that
the budget "won't even pay for the artwork."
(Pike is exaggerating the situation by a factor of about 2. I am not
aware of any single NASA program costing more than about $3B that
produced only artist's concepts. X-30/NASP cost about $7.5B in current
dollars, and part of one X-30 fuel tank was actually fabricated.)
So it isn't any surprise that the plan for Project Constellation calls
for a glacially slow development schedule. The first manned CEV test
flight (of the command module only?) Is scheduled for 2014, ten years
from now. In that time we once developed four manned spaceships:
Mercury, Gemini, Apollo CSM and Apollo LM. Now the best we can hope for
is one (or maybe 1.3 if the rumors of a JSX-style fly-off competition
between two contractors' boilerplate command modules are true).
Furthermore, there has been little discussion about the other key
element of Plan Bush: a heavy-lift booster capable of sending an
enlarged CEV to the Moon. All of the artist's concepts for the lunar CEV
show the Command Module and Service Module being landed directly on the
Moon by a third descent Module, with no equivalent of the Apollo Lunar
This makes sense if the CEV is to be eventually used to exchange crews
doing long tours at a Moon base; it would be dangerous to leave the
command module unattended in lunar orbit or L-1 for long periods. (For
one thing, neither of these locations is stable against perturbations
for very long.)
Previous studies have shown that a minimum moon mission using direct
landing would require a starting mass in LEO of about 1.5 Saturn V
payloads. If we want to launch this in one load, the CEV booster will
need a performance similar to the Saturn VIII/Nova that was planned for
the direct-ascent version of Apollo, or the "Comet" booster proposed for
the similar First Lunar Outpost program that was studied in 1992).
People who say that a manned moon mission could be assembled in LEO out
of small pieces launched on existing boosters like the new EELVs are
dead wrong. This option was never seriously considered by either the Red
Team or the Blue Team back during the Moon Race. It vastly magnifies the
chances of failure.
Both Delta 4H and Atlas 5H can lift about 20 tons to LEO, so many
launches would be needed for each moon flight. The need to design the
moonship in many small pieces increases its total weight. Rumor suggests
that the actual number coming out of current studies of this option are
in the range of 6 to 9 launches (120-180 tonnes). If any one of these
launches were to fail, the whole mission plan would be disrupted.
Also, there is no way we could produce the number of Delta 4H or Atlas
5H boosters it would take to support a serious moon program on top of
all other launch requirements. Since each Heavy EELV uses three core
stages in parallel, 18 to 27 stages would be dumped into the Atlantic
for one Moon landing.
The new factories for these boosters were each designed to produce about
20 core stages per year. So if either of the EELVs was turned over
entirely to the manned program, it could support at best one Moon
mission per year.
A probable compromise plan would be to launch the CEV and its Earth
escape stage in only two launches of a new Saturn V-class booster and
dock them automatically. This plan was considered for Apollo before
Lunar Orbit Rendevous was adopted. Still, it will be very complex and
So where is the money in the Bush Plan to develop "Saturn VB"? Well, it
seems to be off in the nebulous area of the budget chart after 2010. At
this point the Shuttle money becomes available and we at last have real
money available to develop stuff (plus that marching army of Shuttle
servitors with time on their hands). If you add up the big blue budget
wedge, the new program is scheduled to get about $170B between now and
2020. This actually is starting to be a comparable amount of money to
But the Bush Plan isn't just to repeat the Apollo exercise of planting
flags on the Moon. The first goal of the Bush Plan is "return to the
Moon TO STAY." This implies a permanent base, probably at the South Pole
of the Moon where it can tap the ice deposits in the permanently
shadowed zones. This project is somewhere off the left of the current
budget chart in the neverland beyond 2020.
Plans for this moonbase don't take into account the real environment of
the Moon. For instance, official artwork for both Plan Bush and ESA's
similar fantasy program "Aurora" persistently show heroic lunar
explorers living in flimsy aluminum tanks (rather like ISS modules) or
even inflatable plastic tents (rather like the cancelled TransHab). It
seems that people have been working on ISS so long they can't even
imagine anything else.
Remember that big solar flare storm back in November 2003? That event
would have quickly killed anybody in an inflatable hab module or even a
3mm aluminum tank. You need to be under at least 2 feet of moon dirt to
survive even one major flare, and you might have as little as 20 minutes
of warning time to get dug in. And during solar maxima these killer
flares happen about twice a year. So any Moon base has to be
underground, right from the start. The first thing you send to the moon
will be a mobile backhoe to dig in all the other modules as they arrive.
The second thing will be a nuclear reactor to power the base since solar
power won't work in those shadow zones at the poles. Then we need mobile
ice mining vehicles, a plant to split the water, a cryogenic tank farm
to store the H2 and O2, a refuelling vehicle to carry the stuff to the
rocketships.... It will take a lot of flights by Saturn Vbs before our
lunar filling station can start making later flights cheaper.
When you go through this exercise of designing the minimum moon base
that can usefully support further space activity, you quickly come up
with something is actually more complicated than ISS, which we are
unable to support with the combined efforts of the whole world's space
agencies. It will cost about 3 times as much per ton to lift this base
to the moon than it did to lift the ISS to orbit. Possibly we can gain
back this factor of three by using EELV technology instead of leftover
The Augustine Commission said that their similar proposed space program
would require that the then-larger NASA budget be further ramped up at a
rate of %10/yr to a steady level of $40B/yr. There hasn't been any
technological revolution since then than would change this.
Plan Bush as originally leaked to the press only called for a %5/yr
ramp-up in NASA funding. Plan Bush as actually announced has a %5/yr
ramp for only two years, and then reverts to the Clinton-era pattern of
cuts in real dollars. NASA is being expected to carry out John Kennedy's
space program with Bill Clinton's space budget.
Some people say that the answer to this dilemma is a management plan
they call "go-as-you-pay". In this scheme, we restrict our ambitions to
the level of funding that is actually available. If money runs short, we
stretch out the program into later fiscal years and trade-off lower
cost/year for more time to project completion.
"Go-as-you-pay" is just a new marketing name for the existing insane way
that NASA does things. The huge cost overruns on ISS were paid for by
reducing the Shuttle flight rate from 7/yr to 4/yr. The money and
personnel made surplus were mostly transferred to the Station program.
But the lower flight rate led to long delays in the launch of the
Station's innumerable components.
The result of this has been to greatly increase the final cost of ISS,
and insure that the first modules launched are already wearing out
before the Station is even completed. Station is also starting to run
into the problem that almost killed F-22: suppliers no longer make or
service some components which are now obsolete and unsalable in the real
"Go-as-you-pay" ignores the fact that there is a minimum efficient rate
of progress in any program. There is a certain irreducable overhead in
management and facilities that must still be paid for, even when actual
work has been cut to the bone. If we are actually crazy enough to
deliberately use this system in Plan Bush, the date of the first Moon
landing could be pushed back well beyond the goal of 2015-2020 -- and
the cost could increase significantly.
Already, the long time scale of Plan Bush is attracting criticism. It is
very long relative to the average politician's career -- or even the
average engineer's career. Will college students of 2004 really be
inspired to take lots of boring, difficult math and science courses by
the prospect of a Mars mission that won't happen until they are too old
to participate in it???
Clearly, someone at NASA HQ has added up the same numbers I have, and
has decided that a Moon Base is impossible. My scientist friends have
already been briefed by the Space Lords that there will be no Moon base
in Plan Bush, and in fact no manned stays on the Moon longer than 14 days.
This will bring the costs down to the level of the stillborn "Apollo
Applications' program planned in the late 1960s -- but it sure doesn't
sound like a program that would give us adequate experience for a Mars
mission. And it is certainly not what most people have been led to
expect by all those artist's concepts. How much will the total Moon-Mars
program cost? People make fun of the $866B estimate that NASA provided
for a Moon-Mars program in the infamous "90-Day Report" of1989, but if
you study the detailed critiques of this number, you find a lot of
errors and misconceptions.
Critics of the 90-Day Report claim that the program outlined in that
report was a wish-list of programs which mostly did not relate to the
manned Mars mission. NASA shifted a variety of existing and planned
missions from other programs into the Mars plan. An example of this was
the Lunar Polar Orbiter, a pure science mission that was arbitrarily
declared a lunar-resource mission.
But the critics don't seem to have noticed that Plan Bush contains the
same trick. The Jupiter Icy Moons Orbiter and its Prometheus nuclear-ion
engine was already in the budget plan under the planetary science and
technology development budget lines. It has now been moved over to the
new Office of Exploration and its budget wedge is part of Code T's
wedge. Similarly, a variety of ill-defined unmanned missions to Moon and
Mars are to be funded out of that wedge.
Another aspect of the 1989 estimated budget that gets a lot of derision
is the %55 cushion that was added on top for contingencies. The critics
charge that this was a deliberate attempt to puff up the costs so that
Congress would kill the program, and leave NASA in peace to putter in
its sandbox with Shuttle and Station.
But some of these same commentators have attacked NASA over the years
for low-balling cost estimates for major programs, and not allowing
sufficient contingency money! Even simple unmanned probes like the
Discovery planetary missions sometimes run over-budget by %15-20, even
though there is a long experience base for these missions, and elaborate
cost models are employed. It is not unreasonable to expect that costs
for a Moon Base + Mars Landing program might run to %150 of the original
model cost. We simply don't have the experience in this area to be sure.
Ex-Congressman and aerospace lobbyist Bob Walker (now on the Aldridge
Commission that is supposed to oversee the execution of Plan Bush) has
pointed out that when extrapolated out over 30 years the existing NASA
budget actually sums up to about $500B . This date of 2035 would seem to
be the likely time for a first Mars mission under the "go-as-you-pay"
scheme. So the most fervent and extreme critics are only exaggerating
the likely final cost of Plan Bush by a factor of 2or 3.
Plan Bush just proves what a lot of us space critics have been saying
for years: The whole idea of a big government space exploration program
based on 1950s throw-away ICBM technology is an obsolete relic of the
Cold War. Anything really useful in space will cost too much and give us
too little, if it is done in this mode.
But I can't blame the President and his advisors much for this plan. The
Columbia accident brought a huge amount of hidden frustration and anger
with the current pointless manned space program out into the open. Not
proposing some new plan was not an option. Asking Congress for enough
money for a real space program was not an option.
And the kind of radical changes at NASA that would be necessary to bring
the costs down would have upset too many political apple carts in an
election year. The range of politically sellable plans was extremely
small, and Plan Bush as announced is just about in the middle of that range.
However, 2005 will not be an election year. There will be a window of
opportunity to change this unworkable program into one that might have a
chance of sucess (especially if the Shuttle program collapses in early
2005 as now seems possible). In a future column I will outline what this
program might look like.
Jeffrey F. Bell is Adjunct Professor of Planetology at the University of
Hawai'i at Manoa. All opinions expressed in this article are his own and
not those of the University.
More information about the FoRK