Friday, 5 July 2013

Horses fork horses!!!

Mama said papa ain't no negroid, he sure weren't, his mama weren't getting them beers though...

The End Of US Manned Spaceflight Looms Ever Closer

The US manned space program has at best only a few more years of missions left in it, until its cost, complexity and design flaws results in another failure that grounds all US manned launches until a new transport system is designed and built.

by Jeffrey F. Bell Honolulu - Jul 10, 2003  

Once again, NASA has proposed to develop a replacement for the troubled Space Shuttle. This year's project goes by the ungrammatical moniker "Orbital Space Plane". An interim version of OSP called the CRV (Crew Rescue Vehicle) to be developed by 2010 will take over the International Space Station lifeboat task now done by Soyuz.
An improved OSP called the CTV (Crew Transfer Vehicle) will assume the ISS crew exchange task now done by Shuttle in 2012. To minimize development costs, the OSP will be launched on one of the new EELV family of expendable boosters, Delta 4 or Atlas V.
Sound familiar? It should. The OSP is only the latest of many "Shuttle replacement" programs that have all failed dismally. A close look at OSP shows that this program is also doomed to failure due to fundamental technical defects. It's no surprise that such usually reliable NASA boosters as "Space Coast" Congressman Dave Weldon and aerospace lobbyist Lori Garver have publicly attacked OSP.
Most critics have focused on the suspiciously low development costs, or the embarrassing gap between 2006 and 2010 in which no ISS lifeboat is planned. In fact, the basic concept of the program is so stupid that every knowledgeable person involved in it must be perfectly aware that it will never fly.
The basic problem is that the OSP, as currently defined, must carry such heavy mass penalties in the form of wings, wheels, and various escape systems that its performance will not be much better than the Dyna-Soar design of 40 years ago.
Because it cannot carry any of the supplies needed to sustain its passengers once they arrive at the ISS, it will not reduce the number or expense of Shuttle missions needed to support the International Space Station, and will not provide "assured access to space" as NASA claims.
Instead OSP will force NASA to simultaneously fly two very expensive man-rated vehicles at a time when it is financially unable to support even one, and will double the risk of long stand-downs in ISS operations due to lack of either replacement crewmen or the supplies needed to keep them alive.

The Shrinking Spaceplane Mystery:

 The original OSP concept envisioned a 7-seat vehicle which could rescue or exchange the entire ISS crew in one sortie. (NASA's proposed Budget Amendment of 14 November '02 said "as many as 10".) The Level I requirement document reduced this to "at least 4" persons.
This major decline in the OSP's basic performance measure was widely criticized. Although I have not seen an official justification for the 4-seat requirement, it appears to be based on an agreement among ISS users that NASA will be responsible for escape and exchange only of the non-Russian ISS crew members, with the RSA continuing to support 2 or 3 Russian crewpersons with 2-3 Soyuz TMA flights per year.
However, a later NASA document "interpreting" the Level I requirements (online reference) has gone mostly unnoticed. In this 'interpretation" the requirement for "at least 4" seats in OSP has been changed to a "system requirement" that can be reached using multiple spacecraft instead of only one! Presumably, proposals for 2-seat or even 1-seat spacecraft would be now considered acceptable under this bizarre "interpretation" of the "at least 4" requirement.
I know of no other aerospace program in which the basic performance goal has been lowered by a factor of FOUR in the first few months! This isn't just a question of being "a step backward from Shuttle" (or even from Soyuz), but fundamentally wrecks the economics of the program. Even in the CRV mode, a 2-seat OSP is an extremely dubious proposition.
The normal configuration of the station would then be one in which two OSPs and a Soyuz would occupy three docking ports, oriented in such a way that all three lifeboats could be manned and pull away from the Station in any desired order, while leaving other ports free for CTV or supply vehicle docking.
In the CTV mode, the 2-seat OSP would be heavily burdened by the irreducible overhead of basic nav, comm, and docking equipment that cannot be scaled down. So by cutting the seating in half, NASA has much more than doubled the annual cost of rotating ISS crews.
NASA has not given any reasons for this extraordinary lowering of the bar that the three competing contractor teams have to reach. The most likely explanation is that preliminary studies have revealed a 7-seat or 4-seat spaceplane turns out too heavy to be launched on Delta 4 or Atlas V, when all necessary requirements are met.
To see what kind of problems they may have found, let's compare it with the previous, now-cancelled design for a 6-seat Station CRV, the X-38. The X-38 was very narrowly tailored for the CRV requirement. It lacked most of the systems needed for independent flight, since it was to be carried into orbit inside a Shuttle and docked to the ISS with the aid of the Canadarm2. The ECS supported 6 persons for only 9 hours, the RCS used compressed nitrogen, avionics were highly simplified, there was no rendezvous and docking gear, landing used a simple solid retrorocket, parachute and skids.
There was no question of reusability since it was an emergency lifeboat, and since it would only be used once or twice in the lifetime of ISS high reliability was unneeded.
Now let us imagine a CTV version of the X-38. Clearly, a lot of stuff needs to be added: radar, computers, control rockets, fuel, instrument panel, a window to look out of, a docking mechanism that can tolerate significant misalignments and shocks, more O2 and N2 tanks, more CO2 scrubbers, real thermal control, extra batteries.
Many of these systems need to be duplicated to provide sufficient reliability for routine flights, and everything needs to be reusable with minimal maintenance between flights. There just isn't volume for this stuff in X-38 (or any winged vehicle of its approximate size and weight) without throwing out some of the seats.
Of course the controlling factor in all space operations is mass. To see how bad the mass problem is, let's look at the rich and depressing history of previous unsuccessful orbital spaceplane designs:

Table I: Actual and Proposed Space Station Ferry Vehicles
(Ballistic Capsules in Italics):

Project     Designed                 Capacity:   Masses:                  R&D Cost
Name        by    in    Booster      Crew+Cargo  Landing/Launch/ +LES     (FY02 $)
---------   ----- ----  ------------ ----------  -----------------------  --------
X-20 D-Soar USAF  1963  Titan IIIC   1 +  450kg            5165/  6525kg
Gemini      NASA  1964  Titan II     2 +    0kg            1910/(no LES)
Big Gemini  USAF  1967  Titan IIIM   9 + 2500kg                / 15600kg

Shuttle     NASA  1981  Shuttle      7 +12500kg                             $27B
Hermes      ESA   1984  Ariane       6 + 4600kg           15000/ (no LES)   $2.4B
Hermes      ESA   1987  Ariane 5     3 + 3000kg           21000/ (no LES)
Hermes      ESA   1991  Ariane 5G    3 + 3000kg           23000/           $10.1B
HOPE        NASDA 1987  H-2          4 + 2000kg    13000       / 22000kg    $4.9B
HL-20       NASA  1997  Titan III    8 +    0kg           11600/ 16300kg 
HL-42       NASA  1997  NLS          4 + 4300kg    13365/ 21093/ 28725kg
OK-M        USSR  1986  Zenit        2 + 2000kg    10200/ 15000/        
Zarya       USSR  1986  Zenit        6 + 1500kg    12000/ 15000/
X-38+ CTV   NASA  2002  Titan IV     6 +    0kg                /~16000kg

Merkur CRV  USSR  1975  Proton       3 +   50kg            4250/
X-38 CRV    NASA  1996  Shuttle      7 +    0kg     7300/  8163/(no LES)    $0.5B
X-38 CRV    NASA  2002  Shuttle      6 +    0kg            9072/(no LES)    $1.5B
Apollo CRV  NASA  1967  Shuttle      6 +    0kg            4500/(no LES)
Soyuz TMA   USSR  1967  Soyuz        3 +  350kg     2900/  7150/
Progress M1 USSR  1978  Soyuz        0 + 2230kg      -- /  7150/(no LES)

The real killer in Table I is the column labeled "+LES"; these are the total launch masses inclusive of a Launch Escape System capable of boosting the spaceplane quickly away from an exploding booster during max-Q.
This is an invisible element in most spaceplanes, usually tucked away in an "adapter section" between the spaceplane's tail and the top of the booster. Since this adapter/escape module is dumped immediately on reaching orbit it is often not included in the vehicle's "total mass".
However, in those designs where I have been able to isolate its contribution to the total launch mass, it is on the order of %20-30! (The classic "escape tower" used on most ballistic spacecraft is also surprisingly heavy, but it usually is jettisoned after max-Q, so its entire weight is not subtracted from the payload.)
The performance penalty is so great that many spaceplane designers have tried to recover some of it by mounting the escape rockets on the outside of the spaceplane or the adapter and firing them during every ascent after they are no longer required for escape.
This introduces a host of other problems, the worst one being that an extra failure mode is introduced into every launch. The European Hermes mini-shuttle omitted escape rockets completely, relying only on ejection seats even after the Challenger accident.
Since increased crew safety is allegedly a major reason for OSP, it is inconceivable that it will not incorporate a full-capability LES. The huge performance penalty of carrying this heavy module all the way to orbital velocity is the main reason that a 6/7-seat winged CTV cannot possibly be launched on a medium-lift booster like the Delta 4, and even a 4-seat version would be marginal.
In fact, shortly before X-38 was cancelled, a modified version was considered for the CTV requirement, and the proposed boosters were Ariane 5 and Titan 4, suggesting that the project engineers expected the 9-tonne CRV version to bulk up to ~16 tonnes when upgraded to perform the CTV function and fitted with a LES module. So OSP cannot merely be "X-38 on a stick"; it is a different and much heavier beast.
Another lesson from the dreary history of orbital spaceplanes is that the R&D costs are usually underestimated. The Hermes Euro-OSP quadrupled in cost over seven years, and X-38 tripled in six. The idea that one can design and test a new manned vehicle roughly half as complex as Shuttle with a budget only %2-5 as big is clearly a fantasy.

The tyranny of wings:

Several semi-ballistic orbital ferry designs are included in Table I to show just how much we are paying for the dubious benefits of wings and runway landing. One of the main reasons X-20 Dyna-Soar was cancelled in 1964 was that Gemini provided the same capabilities on about one-third the total launch weight.
The Soviet designs OK-M and Zarya, 2-seat winged and 6-seat ballistic CTVs designed to fit the same mass limit, show the same factor of three. A modified Apollo CM was proposed to meet the 6-seat CRV requirement on about half the weight of X-38. Given today's huge launch costs, what possible reasons exist to justify launching two or three times the necessary mass? Many years of Shuttle flights have give some people the idea that reusable spacecraft must have wings, but in fact the only reason the Shuttle has wings is a long-forgotten USAF requirement.
It is perfectly feasible to put a new ablative heat shield on a semi-ballistic vehicle and reuse it. The Gemini 2 capsule was actually refurbished and reflown in 1967 as part of the Air Force MOL program. The Chelomei Design Bureau in the USSR also reflew several examples of a fully reusable 3-man ballistic space station CRV called "TKS-VA" or "Merkur" in 1977-83.
Another myth is that a water landing would require borrowing a carrier battle group from the Navy. For regular scheduled CTV landings near KSC, NASA could use its two dedicated recovery tugs which lie idle at Port Canaveral between the occasional Shuttle SRB recovery missions.
Apollo missions regularly landed within 2nm of the predicted point, so it should take less than an hour to hoist the spacecraft aboard and hose it off with fresh water. For emergency CRV landings, existing search and rescue organizations would be adequate.
The feasibility of a ballistic design for OSP was demonstrated by ESA in 1998, when they flew and recovered a prototype Station CTV called "ARD", which was an %80-scale Apollo CM with modern avionics and recovery gear. Curiously, NASA recently completed a study (online reference) of an Apollo-based OSP design, which does not mention either Gemini 2R, Merkur, or ARD, but instead repeats all the standard anti-ballistic myths.
This is another example of the fact that airplane pilots, who all have a gut feeling that the ballistic spacecraft concept was an unfortunate diversion from the "correct path" of gradually developing airplanes into spaceships, dominate NASA's manned program. (Actually, Table I suggests that this approach makes as much sense as gradually developing steam locomotives into airplanes.)
Although Sean O'Keefe has said that ballistic designs are acceptable in the OSP competition, it is unlikely that any of the three industry teams will propose one. They have received plenty of hints from lower-level NASA pilot-officials, pilot-astronauts, and even some pilot-Congressmen that only a winged, streamlined, Shuttle-like design with sticks and rudder pedals will satisfy them.

The Space-Tech Vacuum.

"But surely all the technological advances made since Dyna-Soar/Shuttle/Hermes will allow OSP to be much faster/better/cheaper," some of you are saying. Whenever anyone says this, I demand that they name those technological advances. Nobody is ever able to, since there haven't been any since about 1964, when NASA's narrow focus on the Moon Race caused them to stop funding basic rocketry research.
What little progress has been made is the gradual reduction in the cost and failure rate of expendable boosters, demanded by and funded by the comsat industry and the DoD. If you look at the current technological shelf that the OSP design teams can pull components off of, it has pretty much the same stuff on it that the Dyna-Soar team had in 1964. (The X-38 did employ the most advanced technology now available, and one can see from Table I that no major improvement in performance resulted.) And it is just not possible to propose to develop anything new within the cost and budget constraints of the program.
Of all the Shuttle replacement programs, it was the ones that tried to develop new technology (X-30 and X-33) that failed most spectacularly, and the one that stuck with low tech (DC-X) that actually flew. New technology in an area as specialized as space flight just doesn't appear; it requires years of sustained effort by large numbers of scientists and engineers at a cost of billions of dollars.
And NASA has not been willing to spend billions of dollars on anything except the "operational" programs, Shuttle and Station. This is the reason our astronauts are flying in Russian capsules and Atlas V will be launching our satellites with a Russian engine. Until NASA makes a major shift in its priorities from current operations to long-term research, don't expect any new technology like aerospike or tri-propellant engines to arrive.

The feeble EELV:

So if a full-sized winged OSP would weigh at least 16 metric tons, which of the many proposed versions of the two EELVs are powerful enough to handle it? Another table is useful here:

Table II: Possible EELV booster configurations
(existing boosters in italics for comparison):

Name of           # of     Cost per            Payload to
Booster           engines  Launch              LEO / ISS / GTO
----------------  -------  ---------           ------------------
“Single-barrelled” medium-lift versions:
Delta 4M          2         $90M(FY99)         8600/ 8400/ 3900kg
Atlas V 501       2                           10300/ 9900/ 4100kg
Delta 4M+ (5,2)   2+2SRB   $100M(FY90)        10300/ 9900/ 4350kg
Delta 4M+ (4,2)   2+2SRB    $95M(FY90)        11700/11400/ 5300kg
Atlas V 511       2+1SRB                      12050/11700/ 4900kg
Atlas V 401       2         $77M(FY98)        12500/12100/ 5000kg
Delta 4M+ (5,4)   2+4SRB   $110M(FY90)        13600/13100/ 6120kg
Atlas V 521       2+2SRB                      13950/13500/ 6000kg

Ariane V          2+2SRB   $180M(FY00)              16000/     kg
Atlas V 531       2+3SRB                      17250/16700/ 6900kg
Titan IV          5+2SRB   $400M(FY97)        17700/17200/
Atlas V 541       2+4SRB                      18750/18200/ 7600kg
Atlas V 551       2+5SRB                      20050/19450/ 8200kg

“Triple-barrelled” heavy-lift versions:
Delta 4H          4        $170M(FY99)             /23800/13130kg
Atlas V HLV       4        $170M(FY98)        25000/24250/12650kg

This table reveals that the two competing medium-lift versions of EELV are not really interchangable, even though they are similar in size and appearance. The basic kerosene/LOX core stage of Atlas V is much more capable than the LH2/LOX core stage of Delta 4, due to the lower energy density of liquid hydrogen.
Delta 4M could not even lift the basic ~9100kg X-38 CRV vehicle, without the extra systems and modules needed for the CTV mission. When augmented by the maximum numbers of strap-on solid boosters, payload to the ISS orbit of Delta 4M+ is only 13100kg while the similar configuration of Atlas V can tote 19450kg.
So to maintain any semblance of competition between Boeing and Lockheed-Martin in producing the manned versions of these boosters, the total launch mass of OSP, complete with all adapters, must be limited to 13000kg, not the 16000-17000kg that was the limit for Hermes or X-38+.
In fact a more desirable goal for both cost and safety reasons would be to launch on the purely liquid-fueled versions of both boosters. Could it be that the rapid decline in the seating capacity of OSP is an attempt to meet this goal -- or maybe to cover up the fact that some senior NASA official made a dumb mistake in judging the launch mass needed?

Launch safety issues:

A simple way to make OSP bigger would be to launch it on the "heavy-lift" or "triple-barreled" versions of Delta 4 and Atlas 5 which can lift about 24000kg.
In fact, many NASA graphics do show the OSP mounted on these much more capable boosters. But this option would raise both the cost and the launch failure rate. The OSP Level I Requirements includes an "increased safety" requirement for the survival of crews, but this is irrelevant.
The basic limitation on the operational lifetime of Shuttle, OSP, or any reusable spacecraft is not the loss rate of crews, it is the loss rate of spacecraft.
Astronauts, after all, are easily replaceable. The number of overqualified applicants vastly exceeds the demand. But the OSP vehicles will be expensive, hand-built national treasures that simply can't be thrown away.
Just imagine what would have happened if the Shuttle fleet had actually flown the advertised 50 times a year -- at a loss rate of 1 in 60 flights, we would have run out of Orbiters long ago. The same logic applies to OSP, only more so because Delta 4 and Atlas 5 are cheap, non-man-rated commercial boosters whose reliability goal is only 98%.
Furthermore, Delta 4H and Atlas V HLV are both likely to have launch failure rates about double those of their medium-lift versions. Experience has shown that reliability of boosters scales directly with the number of stages and the number of non-redundant engines on each stage.
Both heavy-lift boosters have bad configurations from this perspective. The 1st stage is made up of three engines fed from independent tanks, so even a non-catastrophic shutdown of any engine is non-survivable. Both EELV-Hs are effectively 4-stage boosters from a reliability perspective.
Now the old 2-stage versions of Delta and Atlas have a combined recent failure rate of ~%1.6, consistent with the rule of thumb that any individual stage fails a little less than %1 of the time.
This implies that the Delta 4H and Atlas 5H will fail on about %3 of launches, three times the rate of Shuttle. Even if the LES rockets the OSP away from the booster blast, it is left gliding down toward the Atlantic ocean with approximately the subsonic L/D of a brick. Many studies of spaceplanes have shown that they can't survive high-speed high-AOA water landings. Clearly we can save the crews with ejection seats (more weight and volume lost!), but probably not the vehicles.
Now the baseline requirement for the CTV is to relieve the 4 non-Russians on the ISS every four months. So, we have a choice of launching 6 times a year with 2-seat CTVs and splashing one every 10 years, or launching 3 times a year with 4-seaters and... splashing one every ten years.
Apparently, NASA has decided to minimize its losses by choosing the 2-seat option. This is consistent with the announced plan to fly Shuttles with 2-man "kamikaze" crews.
There has already been discussion of reducing the failure rate by developing a special "man-rated" version of the EELVs. This is unlikely to work. A vast amount of effort was put into "man-rating" the shuttle, and the overall failure rate is still %2. And the whole basis of using an expendable booster is that they are cheap.
The basic Atlas V costs only a little more than the $60M Shuttle external tank! Introducing special safety requirements for the OSP boosters will run up the expense and introduce many operational complications.
Others have suggested providing a "Return-To-Launch-Site" abort capability in the OSP which would allow it to fly back to Florida instead of splashing in the Atlantic.
Of course this isn't easy -- otherwise we would already have it with the Shuttle. Any reasonable spaceplane design just doesn't have the gliding range to get back to KSC or even the East Coast after an abort, after it makes the needed high-speed turn. RTLS abort means burdening the OSP with turbojet engines and fuel for a powered atmospheric cruise.
Again, this option was considered for Shuttle Orbiter and rejected due to severe weight penalties and the complications involved in protecting the turbojets during reentry. But the cut in required seating to 2 may be a way of letting the contractors explore the costs of this option.
Another safety problem specific to winged OSP concepts is that the return vehicle's wings are at the wrong end of the launch stack and make the combined vehicle aerodynamically unstable, like an arrow with its feathers at the front. Von Braun's 1952 Ferry Rocket used massive, draggy tailfins to restore static stability, as did the Air Force's X-20A Dyna-Soar spaceplane of 1963.
Nowadays we have mastered the art of flying unstable aircraft like B-2 with automatic control systems that quickly counter any tendency to swap ends. This principle could be applied to the EELV+OSP stack -- if the engines can be gimbaled far enough and quickly enough, and if appropriate control software is written and debugged.
In either case, the EELV used to launch OSP will be considerably different and more expensive than the current version. (NASA has chosen a third solution for the X-37 in deciding to launch the vehicle inside a standard cylindrical payload shroud. This would be unacceptable for a manned vehicle because it would delay escape from an exploding booster.)
For all these reasons, it is clear that an acceptable level of crew safety can only be achieved with a semi-ballistic water-landing vehicle. But I don't have to rely on hypothetical arguments here - the dismal quality control in the Russian space industry has given us living proof in the large number of Soyuz passengers who have emerged bruised but alive from a horrifying series of mishaps that would have killed them if they had been in a "flying Ming vase" like the Shuttle or a winged OSP.

Ground ops and scheduling:

Little attention has been given to the problems raised by having the manned program share launch facilities with one (or both?) of only two US satellite boosters.
The EELVs will also be handling a variety of high-priority military payloads, and also carrying the US flag in the highly competitive commercial launch market. Right now in the aftermath of the Telecom Bubble, Boeing and Lockheed are desperate for government launches to fill up their suddenly empty commercial order books. But the situation might be very different in 2012-2020.
The demand for 6 Medium or 3 Heavy EELV boosters per year, probably in a special man-rated configuration, might prove onerous. Currently, the Russian factory making the RD-180 engine for Atlas V is contractually obliged to deliver only 10 engines per year through 2012, suggesting that the whole EELV program is scaled to a total of ~20 Medium or ~6 Heavy launches/year.
And if Shuttle-like extra safety requirements are imposed, there are sure to be many delays in OSP launches, tying up the pads for long periods. I forsee the EELV program becoming as overstrained as Shuttle was in the period just before 1986 when NASA, DOD, and commercial payloads were stacking up in hangars -- and we all remember what that schedule pressure led to.

Orbital Debris:

Little attention has been given to the problem of orbital debris impact on the CRV while docked to the station. An unnoticed impact on the OSP's wings, nose or belly during its years of orbital storage could doom the crew during reentry. Studies of the debris threat to Shuttle have shown that its thermal protection is very vulnerable. At the current level of threat, a Shuttle docked to the ISS for 20 years has a ~%70 chance of being disabled by space junk.
While 1 big OSP or 2 small ones will make a smaller target, the debris population is increasing at about %4 per year. NASA is concerned enough about this rising threat that the manned sections of ISS are fitted with thick multilayered "space armor" to protect them from debris up to ~1cm across.
The CRV version of OSP needs to be armored against impacts at least as well as the pressurized portions of ISS itself, and the armor must be quickly jettisoned before an emergency reentry. This feature is especially difficult to incorporate into winged or lifting-body configurations due to their high surface/volume ratio.

The real need:

The OSP requirements concentrate entirely on the CRV and CTV tasks. But there are other tasks essential to maintaining the baseline 7-person crew on ISS which are not mentioned. They include:
A) the general supply task now done by Progress.
B) the water supply task. Even with 3-man crews, Progress M1 supply flights have been unable to supply enough drinking water to ISS (the dedicated water tanks on the Progress M version were replaced by more rocket fuel capacity). The Shuttle assembly flights used some of their surplus mass capability to make up for this deficiency. This water shortage is the main reason the ISS crews have been cut back to 2 during the Shuttle stand-down.
C) the orbital reboost task. The Station is constantly losing velocity and altitude due to air drag on its gigantic solar panels. It must be given periodic pushes by its own thrusters or some attached vehicle. The mass of fuel which must be lifted to ISS gets larger each year as station mass and drag area rise (and as the Earth's atmosphere expands in the later part of this decade due to the solar cycle). Even with the current mini-ISS, the fuel capacity of Progress and Soyuz proved inadequate and Shuttles burning their excess OMS fuel supplied much of the reboost.
It is estimated that the completed ISS will require about 70,000kg/year of supplies when fully manned (Space News, 21 April '03). What vehicles will be available to lift this load?

Table III: Capacity and Flight Rate of Proposed ISS Supply Vehicles

Supply       Funding Total   Launch    Flight Rates:       Up       Likely
Vehicle      Agency  Mass    Vehicle   NASA Plan  Reality  Cargo    Cargo/yr
-----------  ------  ------- ------    ---------  -------  -------  -------
Soyuz        RSA       350kg Soyuz     ]            2/yr     350kg   700kg
Progress M1  RSA      7150kg Soyuz     ] 7-12/yr    4/yr    2200kg  8800kg
HTV          NASDA   15000kg Ariane 5G      2/yr  0.5/yr    7000kg  3500kg
ATV          ESA     20500kg H-2            2/yr  0.8/yr   10000kg  8000kg
Shuttle      NASA            Shuttle      5-7/yr    4/yr   12500kg 50000kg

The maximum emergency flight rates assumed by NASA in column 5 are fantastically optimistic. The RSA has recently averaged 2 Soyuz and 4 Progress flights a year, but is contractually obligated to supply only 2 Soyuz and 3 Progress per year through 2006. The European Automated Transfer Vehicle is budgeted to be produced at a rate of one vehicle every 15 months, and is to be launched on the troubled Ariane 5G booster. The Japanese H-2 Transfer Vehicle is a paper concept and the future of its H-2 booster is even more doubtful.
Even if they proceed on schedule, these new vehicles will never do more than keep the European and Japanese crewpersons supplied with truffles and sushi. Most importantly, the Shuttle flight rate has declined from ~7/yr in the mid-90s to ~4/yr today, while the budget has fallen ~40% and is scheduled to drop further in the current 5-year budget plan. So I have drawn up a more realistic flight schedule in Column 6 and a possible cargo budget for ~2010 in Column 7.
This exercise reveals that %70 of the total cargo (and 100% of the US cargo) delivered to the completed ISS must still be delivered by Shuttle. So it will be necessary to continue flying Shuttle missions at the current rate even after OSP takes over the crew exchange mission in 2012.

ISTP: Inevitable Spaceflight Termination Plan.

NASA's current plan is to fly these supply missions in a "kamikaze mode" with only two volunteer pilots on board. This implies no reduction in the onerous safety requirements on Shuttle operations and no reduction in the huge cost. Indeed, the Columbia Accident Investigation Board is likely to call for even more expensive testing of the ageing orbiters between flights, which will run the costs up even more.
Even if the Shuttles are converted to unmanned operation like Buran and reduced safety standards are accepted, they will be hugely expensive relative to our foreign partners' truffle-cans and sushi-cans. So where is the budget wedge to operate OSP coming from? Imagine NASA in the 1980s trying to operate the Shuttle, while still flying 4 Apollo missions every year.
Furthermore, NASA's Integrated Space Transportation Plan shows continued R&D for a true Shuttle replacement vehicle. There just doesn't seem to be any way these three manned programs can be supported simultaneously without a massive increase in the NASA budget.
But the real danger of continuing to rely on Shuttle as a grossly inefficient medium-lift cargo vehicle is: What happens after the inevitable next crash? The loss of Columbia does not impact ISS assembly and supply in the short term, since this orbiter was overweight and not adapted to dock with ISS.
But in the long term it was planned to upgrade Columbia to full ISS support configuration to cope with the growing supply demand. Another Orbiter loss will bring the fleet down to two, except when one of them is in overhaul (at least %50 of the time). Can the necessary 4 ISS supply flights per year be maintained with a 1.5FTE orbiter fleet? Has NASA even studied this problem?
This analysis leads to the question: what exactly will we gain by developing the CTV version of OSP? Why does this task need to be unloaded from the Shuttle, which is going to be flying to ISS every three months anyway? If the Shuttle is too dangerous for a seven-person crew, isn't it too dangerous for a 2-person crew?
And won't the inevitable deaths of 2 astronauts on a mere cargo-carrying flight which the other ISS partners are all flying with safe unmanned vehicles be just as traumatic at the Challenger and Columbia tragedies, and result in a similar hiatus in vital supply missions to ISS? So OSP-CTV will be grounded anyway, despite its superior (and expensive) safety systems, because its passengers will have nothing to eat or drink when they arrive at the station (except whatever truffles and sushi our gallant allies can spare).
Clearly, the OSP by itself does not provide "assured access to space". The USA needs to also design (or buy) an unmanned spam-can supply vehicle adapted to EELV launch, to replace the Shuttle in its cargo-carrying role.
This need is even plainer now that the true state of deterioration in the remaining Orbiter fleet is becoming clear. Yet no money is budgeted in the Integrated Space Transportation Plan for this vital vehicle! In fact, NASA has terminated funding for the Assured Access to Space program, the only current program which might have produced such a vehicle.

Grasping at straws:

The obvious inadequacies of OSP are inspiring many proposals for "faster/better/cheaper" ways to achieve the program's objectives. Unfortunately, most of these ideas are as ill-founded as OSP. Many of them involve reducing the already inefficient flight rate of the Shuttle, or assigning it new roles that it is quite unsuited for.
Typical of these desperate ideas is a proposal to put Shuttle on "standby" status for occasional special loads. This is a fantasy, since the Shuttle program relies so much on highly skilled manpower that it can't be turned on and off at will.
Another unworkable concept is to use Shuttles as the CRV by leaving them docked to ISS during each crew's 4-month deployment. This proposal would knock us down to a 2-orbiter fleet just like another crash (except it will cost more because there would still be 3 orbiters to maintain).
Furthermore, the large and fragile target that the Orbiter presents to space debris would require that all external surfaces be inspected frequently for impact damage. And if dangerous damage IS found, what do we do about it? The Shuttle lifeboat proposal would merely accelerate the ongoing decline in the Shuttle fleet and hasten the black day when the US loses its manned space capability completely.

What is to be done?

Fundamentally, the NASA manned program today is in exactly the same situation described in a prescient OMB report written back in 1969, during the planning stages of Shuttle:
The critical problem with manned space flight is that no one is really prepared to stop manned spaceflight activity, and yet no defined manned project can compete on a cost-return basis with unmanned space flight systems.
In addition, missions that are designed around man's unique capabilities appear to have little demonstratable economic or social return to atone for their high cost. Their principal contribution is that each manned flight paves the way for more manned flight...
NASA equates progress in manned space capability with increased time in space, increased size of spacecraft, and increased rate of activity. The agency also insists upon continuity of operational flight programs, which means we must continue producing and using current equipment concurrently with development of next generation systems. Therefore, by definition, there can be no progress in manned space flight without significantly increased annual cost.(Quoted in Heppenheimer, THE SPACE SHUTTLE DECISION)
NASA has repeatedly tried to get out of this self-inflicted trap by conning someone else to develop Shuttle II out of their pocket (X-30: Air Force, X-33: LockheedMartin, OSP: the comsat industry). This idea has failed every time.
In an ideal world, it would be time to try another option: Stop Shuttle flights, stop the Space Station program, and divert the money absorbed by the marching army of Shuttle/Station workers into real research on a real spaceship.
Of course, this would require serious thought and public debate on what kind of spaceships we need, instead of just replaying the obsolete 1952 Von Braun plan over and over again. It would require the Bush Administration to show the same kind of resolution in standing up to our "international partners" on space policy that it has shown on insane-dictator-control policy.
It would require Congress to actually adopt the role of skeptical overseer of public expenditure that it plays in all other areas of government activity. It would require everyone to admit that 14 Shuttle astronauts really did die for nothing.
Before Columbia, these things were off the table. But right now, influential people are starting to consider them. There is, however, one wild card that nobody seems to be talking about: the impending launch of the first Chinese astronauts.
Since many people still think that manned space flight is some kind of measure of national power (thank you, Nikita Khrushchev!) the first Chinese flight will produce another Sputnik Shock and pressure to continue a spectacular US manned program. will be irresistible. So it is more likely that NASA will be allowed to continue assembling the International Space Station at a glacial pace.

A Modest Proposal

Even if we can't get the Space Station millstone lifted from NASA's back, we can at least get it off the present suicidal course of flying the decaying Shuttles around in circles until they have all crashed. I propose the following program:

  • Inspect the Apollo Command Modules currently in museums for heat-shield damage that would prevent them from surviving a second reentry from orbit. The CMs that never left LEO should still have well over 50% of their ablative material remaining. Choose the two best of the lot and refit them as 6-seat CRVs, along the lines of the 5-seat conversion planned for the unflown Skylab rescue mission. (The panel of superannuated engineers that studied this option for NASA raised several objections, but none of them seem to be show-stoppers.) The converted Apollos should be ready by the time the current CRV contract with the RSA expires in 2006, and one or both will be carried up to ISS on a Shuttle flight.
  • Begin negotiations with Daimler-Chrysler for the purchase of a NASA version of the ATV truffle-can cargo ferry, adapted for launch on EELVs. To make the negotiations more competitive, start studies of an all-American spam-can design to meet the same performance goals. Study a tug version of ATV that could deliver ISS modules without risking lives.
  • Change the OSP program specification to eliminate all CRV-based requirements and require that all proposals include 4 or 6 seats and use the proven Gemini or Apollo semi-ballistic configurations, with crew safety assured by an escape tower. Emphasize this by removing "plane" from the program name, and stopping the recruitment of airplane pilots as astronauts. Consider using the weight saved for a disposable cargo module. Allow non-reusable designs if they turn out to be cheaper. Ask RSA and ESA to provide full data on the Merkur and ARD ballistic ferry vehicles.
  • Reprogram the billions of dollars now being budgeted by the Air Force for direct subsidies to the Delta 4 and Atlas 5 contactors towards the purchase of actual boosters for ISS supply missions.

What I can tell you country boy is ship vs ship he was...

McVicar - Roger Daltrey (The Who)