Right Side Up: A History of the Space Transportation System

Chapter 13: Reentry
“Anything beats an expensive stack of paper.”

Chapter 13: Re-Entry

In the face of that coming hypersonic storm, the feeble impulse of the peroxide thrusters would be of no use, and the muscle power of the human pilots would not suffice to move the mammoth control surfaces on the Lifter’s trailing edge. Here, too, the computer would exert its will upon the ship, coordinating a much faster, wilder dance of hydraulic pumps and motors, deflecting air around the ship to guide it down to the regime where the human mind could again respond quickly enough to make a difference.

In truth, the phrase “edge of space” is not just a misnomer but a nonsense. The atmosphere does not end, but gradually gets thinner and thinner, until the very faintest wisps of Earth’s atmosphere merge with the thin stream of gas constantly shed by the Sun. The drag force it exerts on a spacecraft, then, is never truly absent, but only stronger and weaker depending on one’s location and velocity.

Slowly, imperceptibly at first but with ever greater intensity, that force grew as
Constitution plunged back toward Earth, her broad, flat belly forward, shielding the crew cabin and the engines from the heat of reentry. As she fell, the air beneath her couldn’t get out of her way fast enough, and like the piston in a diesel cylinder hundreds of kilometers long, she rammed the air into a smaller, hotter volume. Hypersonic shock waves formed around her wings and nose, stagnating the local air flow and generating even hotter temperatures.

Unlike the Apollo and Gemini crews, or their compatriots aboard
Endeavour, Young and Crippen were only aware of their ship’s assault upon the atmosphere peripherally, through skin temperature gages on their control panels. Constitution was not moving nearly fast enough to heat the air to incandescence, much less to ionize it into a plasma--where earlier astronauts had blazed a glowing trail through the sky, their fall back to earth was understated, calm. Ground tracking cameras had no trouble watching Constitution as she descended back through the atmosphere.

In the thickening atmosphere,
Constitution transitioned from a spacecraft to an aircraft. Her ailerons and rudders once again dug into the ever more substantial airflow, exerting immense forces and bending moments on the great ship. But Constitution came from good stock--her ancestor, the Saturn V S-IC, had been built with large structural margins by men more suited to building locomotives than tight-margin missiles. She took the loads and ploughed ever deeper into the atmosphere, through aerodynamic forces that would have already shredded a lesser rocket…

In the final years of the seventies, the debut test flights of the Space Transportation System had occurred amid major questions about the program: Could a reusable booster really be built and flown with the aggressive semi-retropropulsive, semi-aerodynamic entry and return profile? Could such a booster be effectively reused between flights? How many flights would really be feasible between major overhauls? Would they have sufficiently long lifespans to enable their higher initial costs to be spread enough to beat out the costs of contemporary launchers? Could they turn around fast enough to meet the aggressive flight schedules sold by NASA to Congress, the American public, and institutional and commercial customers? In a little more than a decade, more than a hundred Lifter and Shuttle flights had conclusively demonstrated that the answer to every one of these questions was yes. The Lifter was the vehicle of choice for NASA’s orbital manned spaceflight, for US military reconnaissance, and for commercial payloads. While other launchers like Ariane scrambled for mere tens of payloads, bolstered mainly by European institutional support, Lifter was launching nearly that many commercial communications satellites every year, massing more and carrying more capability than customers could have found on any other system. The USAF had even used Lifter’s massive payload and the Shuttle’s orbital capabilities to demonstrate value for highly classified manned missions to rendezvous with their latest spy satellites in polar orbit. The dreams of Blue Gemini, the Dorian/MOL, and the Dynasoar had come true in the form of the STS-101’s first flight of an all-military crew to space to offer manned assistance--if only in systems maintenance--to orbital reconnaissance.

However, by May 1989, world events offered new questions as NASA and STC moved laboriously to Lifter’s Return to Flight. In the near yearlong absence of Lifter and Shuttle, new questions had been raised. The Space Transportation System had changed spaceflight and in some ways the world, but what role would it find in its second decade in the world it had created? What would be the effects of Magellan’s loss on STS-116 on the shape of policy for manned and unmanned spaceflight? What did Lifter’s first major failure in more than a hundred launches mean for the next generation of space launch systems, both in the US and abroad? Whatever the answers to these questions, though, one thing was sure: the Space Transportation System was and would remain the keystone for Western access to space. While its supremacy might be challenged by new competition from other Western vehicles or by potential access to the Soviet Groza system, that would only come into play in the long term. Thus, while the winds of change saw thousands of engineers, program managers, lobbyists, and politicians debate the future of spaceflight around the world, thousands more had to fight through the winds to focus on returning the Lifter to flight with its new dual-engine upper stage.

The groundwork for the return-to-flight with STS-117 had been laid over the development of the Magellan Review Board’s findings, but the final steps came in the form of the qualification firings of the S-IVD Dual Engine Upper Stage vehicle. After the single failure to light the engine on STS-116 had sent the Magellan probe tumbling to its doom, NASA and STC had gone to the unrequested step of transitioning all future Lifter missions to the more redundant, higher-capacity DEUS. Even for payloads where the S-IVD’s enhanced performance and theoretical improved safety were not specifically required, STC made the decision to switch. Doing so was a critical step to rebuilding their reputation with commercial launch customers, but also simplified production, meaning STC would avoid having duplicated lines for two stages. As Lifter’s flight rate had risen, the question of having two such lines had become more critical: throughput was high enough to depend on repeated production operations and leverage economies of scale, tooling, and expertise, but still low if this production was to be split across two stages with nearly entirely distinct thrust structures, pressurization schemes, and other interfaces. However, it meant that the test program for the DEUS directly paced the return-to-flight for the entire program.

In spite of this, the S-IVD qualification program was extensive, but it was compressed by test engineers working in shifts nearly around the clock to get the initial S-IVD-T qualification stage in and out of various test cells at NASA’s Stennis test site. As results came in while the stage was poked and prodded, rattled and shaken, and finally fired over and over, McDonnell engineers back at Huntington Beach, California worked evenings, nights, and weekends to process it. During the qualification of the integrated stage’s ignition transients and cross-comparison of data with the extensive firing of each of S-IVD-T’s engines before installation, the manager of the engineering team responsible for the Augmented Spark Igniter redesign, which was an area of particular focus, purchased for his team T-shirts bearing the team member’s names and the phrase, “No Sleep ‘Til Orbit”. However, even as S-IVD-T was put through its paces, the first three production S-IVD stages (D-001 through D-003) were on the production stands, as Huntington Beach continued its proven pattern of building stages in three-unit lots. At least for the moment, the practice of hot-firing completed stages on the way to launch was resumed, after having been deleted for schedule and cost control purposes. Before 1988 ended, the S-IVD-001 stage, earmarked for the STS-117 return-to-flight mission, joined S-IVD-T at Stennis for its qualification firings with 002 and 003 not far behind.

The qualification program went smoothly, but there was only so far that engineers could push themselves while remaining confident in their tests, their data, and their analysis. As production and testing procedures were overhauled and the conformance of S-IVD-T and S-IVD-001 were verified, February 1989 melted away. The booster for STS-117, the freshly-overhauled Constitution was already stacked and waiting in the VAB when the first live DEUS was delivered to the Cape on February 23rd. By the time the stage was re-inspected after transit, and final integration preparation was carried out, the launch date had slipped into March. Finally, however, STS-117 roared into the Florida sky on the long-anticipated return to flight on March 14th, 1989. Flight controllers were laser-focused on their data during the count as elevated upper-level winds which had plagued the previous day’s launch attempt threatened to once again violate launch constraints. The winds settled within tolerable limits shortly before launch, and the actual staging and the first ignition of a Dual-Engine Upper Stage in space were picture perfect. The successful deployment of a pair of commercial communications satellites, whose owners had received a major discount from STC to accept the STS-117 launch slot, brought the return-to-flight to a new phase.

With STS-117 down, the problem was to ramp launch frequency back to the levels which had become typical prior to STS-116, with launches routinely occurring twice in the same month. More rapid turnaround from KSC’s twin LC-39 pads was possible, but unnecessary as the flight rate was able to meet the available payloads, particularly with the double-manifesting of communications satellites. However, with Lifter out of service just weeks shy of a year, there were literally dozens of payloads which were either due for launch in 1989, or which had been scheduled in 1988 and had slipped with the STS-116 failure. The build up of the launch cadence started slow: the stacking of STS-118, with the Space Shuttle Discovery and Lifter Liberty, didn’t begin until STS-117 was safely flown. The mission followed in its turn from LC-39B on April 27th, with Discovery’s crew headed to Spacelab to ferry critical supplies and carry out overdue maintenance on the orbital outpost after over a year untended. However, by the time Libertyflew, STS-119 and Independence were already being prepared with another pair of communications satellites. It was less than three weeks later that Independence followed her younger sisters, and STS-120 on May 20th confirmed that STC was back on pace to meet all its obligations. In fact, to catch up on backlogged flights, Kennedy Space Center was to see no fewer than seventeen Lifter launches in 1989, with another three from Vandenberg including Resolution on a LUCID servicing mission using the new DEUS performance. Fresh off the failure of STS-116 and the critique of people like William Proxmire, Lifter had showed its competitors what they would have to match by breaking its own prior flight-per-year record.

With the Space Lifter back in action and STC and their teams working above and beyond to clear the flight backlog and return to a regular launch cadence, NASA had finally worked through enough turbulence to look beyond the day-to-day operations and into future planning. However, the election of George H. W. Bush as the 41st President of the United States had brought massive changes to the way that the American civil space program was run. In order to better coordinate civil, military, and commercial space efforts in the US, Congress had authorized the creation of a National Space Council, answerable directly to the Vice President, and through him the President, designed to “set ambitious goals and maintain American preeminence in space, while further integrating the High Frontier into the American economy,” in response to the embarrassing loss of Magellan and the rising Soviet success of Mir. The activities of American private and semi-private space firms, including Space Transportation Corporation, Geostar, PanAmSat, and other new satellite communications companies, were now also distinct enough from NASA and other US government functions to merit oversight and coordination beyond a mere office at NASA headquarters. It was hoped the NSC could recommend to the President the most effective ways to promote continued American success in space on all fronts, transcending the bureaucratic limits of the civil, military, and commercial sectors.

Bush’s surprising choice for head of the NSC was Mark Albrecht, who had been a Senior Research Analyst working for the CIA on the Strategic Defense Initiative, and who had written the Republican Party’s 1988 platform on defense. Though he had ample experience with the policies and management of the USAF’s space policies, he had not previously worked with NASA, raising some concern as to whether he could actually tackle the challenge of giving the agency a new direction.

President Bush’s nominee for the new administrator of NASA also raised eyebrows. On Albrecht’s recommendation (for he had worked with him on the SDI), Bush nominated a little-known middle-manager at TRW named Dan Goldin, who had distinguished himself by applying advanced microelectronics technology to satellite design, and for pitching a cheaper design for NASA’s Earth Observation System satellites, emphasizing modularity and shared components with commercial satellite busses and the less-classified Department of Defense intelligence satellites. Though competent, he was essentially a “nobody” in Washington--it was, in fact, not until his confirmation hearings that his registration with the Democratic Party became public knowledge (somewhat to Bush’s annoyance, though, as Dan Quayle noted at the time, “he certainly didn’t have any trouble getting confirmed” in the Democrat-held House or Senate).

The third individual who formed the “Space Troika” of the Bush Administration was Vice President Dan Quayle. Like Bush, Agnew, and Johnson before him, Vice President Quayle was expected to handle the NSC’s day-to-day operation and make recommendations to President Bush. It was Quayle who first proposed that Bush should make a major space policy announcement on July 20, 1989, the twentieth anniversary of the landing of Apollo 11 on the Moon. Bush, eager to counter criticisms of his “lack of vision” and possibly in an effort to step out of Ronald Reagan’s immense shadow, readily agreed. From April to July, the National Space Council would work with NASA and representatives from STC and, to a lesser extent, other American aerospace firms to determine the best way forward for America’s civil space program.

The one feature that most united Goldin, Albrecht, and Quayle was a consensus that they had to operate within realistic budgetary restrictions. Conscious of Agnew’s failure to pitch Tom Paine’s vision of a mission to Mars by 1986, Quayle wrote in a memo in late April of 1989 that “the Democrats who control congress are not LBJ. The man in Moscow is not Khrushchev. President Bush doesn’t have a dead predecessor to avenge. Those are our constraints.” In this light, the nominations of Goldin and Albrecht, both innovative, fat-trimming managers with a history of effective cost and scope control, becomes less surprising. With their constraints in mind, and after a series of meetings with upper management at NASA and at STC, and with Norm Augustine at Martin Marietta, the National Space Council (NSC) turned to the recommendations of the National Commission on Space (NCS) and worked to determine which technologies were on the critical path to Mars, what infrastructure would be needed to prove them, and which of those technologies really needed up-front government support.

To this end, the NSC took the NCS’s list of enabling technologies and infrastructures for crewed missions to Mars and began whittling down those deemed less central to NASA’s mission. Based on discussions with executives at TPLI and Martin-Marietta, they concluded that the private sector was already developing lower-cost launch vehicles, making a government-funded one redundant, at best. The assumed near-term availability of such vehicles also reduced the urgency of developing advanced in-space propulsion technologies--if the cost per-kilogram to LEO fell far enough, the importance of reducing initial mass in LEO fell with it. This left as the main technological goals for a human mission to Mars the development of in-space nuclear power sources, a reusable interorbital tug, and a closed-loop, long-term life-support system. While each of these three technologies would require a large research and development effort, none of them in themselves could satisfy the primary goal of President Bush’s planned new direction in space--to demonstrate American preeminence. The American public, and the public overseas, would not see a qualitative difference in the scope of American activities in space if only these technologies were developed or even flight-tested. Satisfying the President’s desire to demonstrate American preeminence would require a near-term goal that could easily be conveyed to the public. Following this train of thought, Goldin and Albrecht summoned a commission of engineers and scientists from the major NASA centers and asked them to design reference missions for a human lunar return by the year 2000, with the caveat that as much of the new technology and infrastructure developed for such a mission be applicable to a Mars mission some time in the twenty-first century. Even as they worked, President Bush made his great speech at the National Air and Space Museum, flanked by Neil Armstrong, Buzz Aldrin, and Michael Collins, the heroes of Apollo 11:

“In 1961 it took a crisis—the space race—to speed things up. Today we don’t have a crisis; we have an opportunity. To seize this opportunity, I’m not proposing a 10-year plan like Apollo; I’m proposing a long-range, continuing commitment. First, for the coming decade, for the 1990s: A new cislunar infrastructure and a return to the Moon, with a sustainable, reusable architecture, building upon our successes with the Space Lifter for the past decade. Next, for the new century, to open the Moon to American industry as Earth Orbit has been opened, to tap the physical resources of the High Frontier. And then, journeys--not just one, but many--beyond the Moon, to the other planets, leveraging again the skills we built on and around the Moon, beginning with a Manned Mission to Mars.”

The hidden genius of Bush’s speech was that it recognized Mars and the other planets as the goal for which his new program aimed, but it left the actual planning for Mars missions until some undetermined point after the technology was refined in cislunar space. Though this approach received some criticism among some sectors of the space advocate community (and from Martin Marietta, whose Vice President for Space Operations would go on to propose in 1998 that all that was really needed for Mars missions was a slight modification of existing launch vehicles and LEO systems), in practice it took a great deal of pressure off NASA’s engineers and managers, as they did not need to design Mars missions to fit an, at best, modestly-increased budget. Indeed, a preserved memorandum from Administrator Goldin to Vice President Quayle indicates that concerns about controlling overall program costs were already surfacing at NASA and the National Space Council in May of 1989, as Goldin warned Quayle that, since the idea of the program was to design hardware that could be modified for Mars missions later on, it didn’t make all that much difference to the final schedule whether the Mars program begins in 1990 or in 2000. Therefore, the memo continues, NASA should focus on pitching the lunar return program first, as it was easier to secure funding for one part of the program than for both, and because such an approach gave the agency and its partners greater flexibility down the line. Goldin made reference to the “phased development” approach NASA had taken to the Space Transportation System, which had yielded the reusable booster, a reusable orbiter (though without its own significant propulsion), and a space station, which had yielded immense benefits for the agency even without the remaining elements of the STS. The fact that the second phase of that development (the large, integral-propulsion reusable orbiter and reusable space tug) had not yet manifested was noticeably absent from the memo.

The architecture that emerged in response to President Bush’s call for a Space Exploration Initiative (as the effort came to be known), developed by engineers from Johnson, Marshall, Kennedy, and STC, with consultation from every prime contractor in the American space industry (and quite a few of the secondary contractors), thus centered on operations on the Lunar surface and in Lunar Orbit. The new architecture called for a reusable in-space transport vehicle (the long-delayed Space Tug) providing logistical support to a reusable lunar lander, which could carry either cargo or crew down to the lunar surface from a small orbiting maintenance platform. The reusable Space Tug would, in addition to servicing the lunar lander and lunar orbital platform, deliver satellites to geostationary orbit and inject probes to interplanetary trajectories, providing a cheaper alternative to the Centaur upper stage and amortizing its development cost over more missions. The technologies developed for the Tug and Lander would also have applications for the long-term storage of propellant for Mars or other destinations.

The proposed program, Option B, was one of three paths forward presented to President Bush in the early autumn of 1989. The other two, Options A and C, called for, respectively, a 20-year ramp-up of space activity in cislunar space and on the Moon culminating in a landing on Mars by 2012, and a lower-intensity program of technology development in cislunar space (essentially, the recommendation of the NCS in 1986). NASA presented President Bush with cost and time estimates for the various milestones of each project, with Option A featuring a lunar landing by 1998 and a permanent base in 2001, for a total price-tag of some $200 billion. Option C was somewhat more nebulous--each component of the program, from a full-time space station in Low Earth Orbit to test out closed-loop life-support technologies to a completed Nuclear Thermal Rocket development program to a new hypersonic flight development program, had its own schedule and cost. What they lacked, in Bush’s eyes, was a concrete end-point at which the United States could declare “Mission Accomplished!”

Option B, while nominally aiming to develop a system that could be used for Mars missions, did not give cost or schedule estimates past the year 2000. It called for the completion of the interorbital Space Tug by 1996, and for lunar landings by 1998. Though the Lander would be of great utility in building a base, that was left to the next administration. Similarly, though Option B also called for a small, full-time space station to serve as a test-bed for “long-term space habitation technologies,” it did not propose schedules or costs for an interplanetary version of this space station. This greatly reduced the cost estimates that NASA could suggest to the President--compared to Option A’s $200 billion price tag, Option B was estimated at just under $40 billion, spread over 8 years. For that price tag, NASA would have three new vehicles (the Tug, the Lander, and the Space Station), an American flag on the Moon again, and a small suite of new technologies that could indeed be directed toward human Mars missions in following administrations. Furthermore, once the Tug and (possibly) the Lander were spun off into a new contracting organization (as STC had been spun off to operate Lifter), the operational costs would (theoretically) fall off and operations between LEO and the Moon would fall to the private sector, just as operations between Earth and LEO had.

By early October, Bush had been sold on Option B, and the Space Troika’s challenge had shifted from the comparatively simple task of briefing a sympathetic President to the much more complex challenge of selling a flashy new technology program to Congressmen already salivating over the fruits of 44 years of Containment…
 
Last edited:
Man, why do the ATL always get the cooler space stuff?

(Although, I guess it's selection bias, given that a TL where the space program was even less successful than our own wouldn't make for very interesting reading...)
 

Puzzle

Donor
(Although, I guess it's selection bias, given that a TL where the space program was even less successful than our own wouldn't make for very interesting reading...)
It could have a world war three where a lack of ICBMs meant that neither side had to hold back as they fought in the ashes of what was once Germany...
 

Archibald

Banned
Man, why do the ATL always get the cooler space stuff?

(Although, I guess it's selection bias, given that a TL where the space program was even less successful than our own wouldn't make for very interesting reading...)

If Musk grand plan ever come true - "cool" - no, "awesome" would be a more appropriate word.
 
How did this person find out about OTL?

Great work as always.

It's actually an OTL quote--though, given the circumstances behind it, it can equally likely be said ITTL, if less cynically.

Glad you like it!

Man, why do the ATL always get the cooler space stuff?

(Although, I guess it's selection bias, given that a TL where the space program was even less successful than our own wouldn't make for very interesting reading...)

Selection bias, and also because this TL is inherently optimistic--we're writing what the STS should have been, in our minds (as the title goes, it's "right side up" ITTL, but upside-down IOTL).

It's equally likely that Nixon could have scaled the program back to Titan IIIMs and manned flights every other year, or Carter could have canned the whole thing entirely. But, as you observe, a TL focused on that would be short and kind of depressing.

I'm glad everyone's enjoying the post so far. Now that Lifter's back online, we'll be moving into meatier content--TTL's SEI, and a look at other advances around the world. Stay tuned!
 
Chapter 14: Turnaround
"We must, indeed, all hang together, or assuredly we shall all hang separately."

Chapter 14: Turnaround


Constitution plunged deeper into Earth’s atmosphere, trading height for velocity and then dumping the resultant kinetic energy overboard as heat, leaving a trail of superheated air through Earth’s atmosphere. Her peroxide thrusters were silent now. Instead, the potent chemical was routed into a turbine, the auxiliary power unit, which provided the force that drove her great hydraulic motors. Those motors pushed on the ship’s enormous rudders and ailerons, cutting into the surrounding airflow and pushing the ship back home. By now she was out of the realm of missiles and target drones, back under Mach 6, in the regime where human reflexes and training could once again react to the changes fast enough to matter.

High above the Atlantic Ocean,
Constitution, still travelling several times the speed of sound, began her long arc back toward Cape Canaveral. She turned to port, toward the north, passing the storm front that had delayed her launch, returning to clear, calm weather over the western edge of the Sargasso Sea. Even from their high, hurtling perch Young and Crippen could have seen, had they a window facing down, how calm the water was here, and made out some of the brown patches of kelp that dotted this region of the Atlantic. Constitution maintained her broad, banking turn, bleeding off more and more speed as she turned back toward the continental US.

As she did so, her pilots kept a close watch on her rate-of-descent indicators, noting that the rate climbed with time, for
Constitution remained, aerodynamically, a brick. She had never been optimized for slow, level flight, but for the job of lifting a payload to the upper fringes of the atmosphere. As her speed dropped, so did the lifting force on her wings, and she acted more and more like the wingless fuel tanks she counted among her ancestors. Unassisted, she would not have managed to return to land, and would have flopped sadly into the Atlantic. This had been obvious to her builders, and they had given her a way to overcome that limitation…

Most historians agree that the start of the Cold War can be dated to 1947, when President Harry Truman announced the Truman Doctrine, pledging to defend any state threatened by Communist expansionism. This doctrine effectively began the US’s general scheme of “containment,” fencing in the Soviet Union and keeping it far away from the US (those countries already fallen to Communism were left to fend for themselves). From that year until 1991, the US labored to contain its main rival, building up a massive military-industrial complex, developing advanced new weapons and delivery systems, stationing troops across the globe, toppling governments, funding rebels, and projecting soft power through, among other things, the civil space program.

Because of its role as one of the most visible faces of American Soft Power (some 450 million people had watched Neil Armstrong take his first steps on the Moon, roughly ⅛ of all humans alive at that time), NASA had always counted on consistent, if not always expansive support from both the Executive and Legislative branches of the American federal government. Unfortunately, with the demise of the Soviet Union and the end of the Cold War, the agency found itself looking for a new justification for its major programs, as American congressmen suddenly discovered a passion for fiscal responsibility and made promises of new social programs (or, more often, tax cuts). Even as the President proposed a sweeping new plan to expand the agency's goals back to the moon and beyond, NASA found itself struggling to justify the funding which such endeavors might require. The agency approached its international partners, the European and Japanese space agencies, in an attempt to expand the Space Exploration Initiative from a program of discretionary spending to one of international diplomacy. While the Solar Polar Mission a decade earlier had shown that that did not necessarily make the program bulletproof, it was hoped that offloading some of the development costs to the international partners would reduce costs and make the program more palatable to Congress, while also making the more foreign-policy-minded Senators more hesitant to cut budgets.

The European Space Agency, drawing on its years of experience operating Spacelab, and its ongoing experience designing the planned Spacelab emergency reentry capsule, the Crew Return Vehicle, was fairly enthusiastic about the opportunity to work with NASA on a return to the Moon. It did not take long for ESA to propose the use of their Spacelab CRV for the lunar program. Though the program’s heritage includes Manned Ariane studies that date back to the mid-1970s, it formally began in 1983 as part of a study into extending the Spacelab space station into a permanent outpost. As the American Space Shuttle did not have the ability to remain docked to Spacelab indefinitely, a new vehicle was required that could remain docked at Spacelab for up to three years at a time, and which could quickly return a sick or injured crewmember to Earth for medical attention.

Though early designs favored a lifting-body or winged reentry vehicle, whose crossrange would enable landings in either the US or Western Europe on a runway, development cost constraints and mass restrictions on Ariane (whose use as a manned launcher remained a long-term goal in certain circles of the French space agency) led CNES to focus on a headlamp-shaped capsule, very reminiscent of the Soviet Soyuz and Zond designs. With a diameter of 2.8 meters, the capsule would have room for four crewmembers in rather cramped seats. Though the seating was not a serious concern during spacecraft design (as an emergency return capsule, by definition, would only be occupied for a few hours at most), the capsule design did limit the spacecraft’s cross-range, forcing the ESA to investigate landing sites outside of Europe and America in order to guarantee that, in very dire circumstances, crews would not need to wait several orbits before landing. Oceanic landings were ruled out for both reasons of time and budget, so the ESA and the French government sent representatives to almost all large countries over whose airspace Spacelab orbited, including the United States, Australia, Brazil, India, the Soviet Union, and large parts of the former French and British empires in Africa (China, oddly, was not consulted for possible landing coordination, and discussions on the topic would not take place until the late 1990s). Ultimately, renewing its historic partnership with the European space ministries, Australia offered its Woomera Test Range as the main backup landing site for CRV capsules.

Through the use of more modern manufacturing techniques and lightweight alloys, and the use of a very minimalistic “service module” (essentially a set of retrorockets with explosive separation bolts), the entire spacecraft came in at under 4,500 kilograms (a further 1,000 kilograms was allocated for a maneuvering bus that, launched with the CRV on an Ariane 2 rocket, would guide it to a docking with Spacelab). It was this design that was frozen in 1985 as “Asclepius,” after the Greek god of medicine.

By 1989, the European CRV was far-enough along that almost every analysis conducted at NASA’s Johnson Space Center found it to be the fastest and most cost-effective path to a lunar return vehicle. While the Space Shuttle was well suited to returning crews from Earth orbit, the higher speeds of lunar entry would subject it to far more heat than it could handle, and the added weight of the Shuttle's cargo bay and winglets would be of minimal use on circumlunar or lunar orbital flights. The European capsule, by contrast, was lighter to begin with, and its simpler shape made uprating its thermal protection system a smaller challenge. Under ESA’s proposal, an enhanced, longer-lifetime, more-durable variant of the Spacelab Asclepius CRV would serve as a crew capsule that would carry the crew to Low Lunar Orbit and back, propelled by a reusable American Lunar Transfer Vehicle. The capsule, with room for 4 astronauts, would separate from the LTV as both spacecraft approached Earth, allowing the LTS to propulsively brake into orbit for reuse. This plan, which European engineers dubbed “Hermes,” achieved a number of goals for both NASA and the ESA--it gave the latter a high-profile contribution to the Initiative, and it saved the former money by making heavy use of a system already in development for Spacelab. The plan, dubbed the "Exploration Crew Return Vehicle" (eCRV) by NASA engineers, was an easy sell to cost-conscious NASA managers, particularly after program managers working on the new Space Habitat program settled on a plan to use the same Hermes system for transits between Spacelab and the new station. Ultimately, many more Hermes than Asclepius capsules would be built, as only the former was planned for actual operational use on either new program. Though the transition from “permanent Spacelab” to SEI would eventually consign Asclepius to that status of a technology development/prototyping program (one prototype was completed, but never meant for flight), the work done on the Block I eCRV massively accelerated the Block II spacecraft’s progress once that program finally received the green light in 1991.

Though Hermes was a popular choice with NASA’s management, it was controversial at Johnson Space Center and with American aerospace companies, who protested the use of a foreign vehicle over their own concepts. Though JSC was eventually brought on board (Goldin’s argument that foreign cooperation helped ensure there was any Initiative at all found a positive reception among the center’s management), Rockwell International and Grumman Aerospace were particularly vocal, the former having not anticipated serious competition with their own proposals, the latter in particular financial pain with the cancellation of further F-14D Tomcat orders and the uncertain future of the Space Nuclear Thermal Propulsion program on which the company’s space division had pinned its hopes. The latter company’s proposal of a cheap MPEM-derived habitat, which would remain attached to the LTS for its braking burn and rendezvous with a Space Shuttle Orbiter, had never been the front-runner at NASA, but its low development cost had had its share of supporters. Ultimately, though, both companies lacked the clout they once commanded. Poor management decisions had left Grumman without a healthy diversity in its portfolio (forsaking high-speed trains and civil aviation, the company later halfheartedly invested in sewage treatment, alternative energy, canoe building, and truck manufacturing, with decidedly mixed results), and it spent the 1990s refocusing on electronic warfare systems, divesting its once-fantastic manufacturing infrastructure. New York’s senators were unsympathetic to the plight of a company that could not even offer many jobs to Long Island. Rockwell International, for its part, did not complain as loudly, hoping that its Rocketdyne division could still secure contracts in the planned lunar lander and lunar transfer infrastructure programs. Though much larger than Grumman, it too felt the pinch of the Peace Dividend, but it still had more to lose.

The decision to use the eCRV as the inter-station transfer vehicle marked the last attempt by Grumman Aerospace to be an independent prime contractor on a space vehicle. Prior to the selection, the options considered by NASA included the eCRV, a new space-only vehicle that would combine an MPEM-derived pressurized cabin with a modified communications satellite bus, and a Space Shuttle orbiter (which would restrict inter-station flights to times when a visiting Shuttle was available). The space-only vehicle would be the most mass-efficient method of travelling between the stations, as its thin aluminum shell did not need to have any protection from atmospheric stress. In many respects, the proposal resembled older Grumman attempts to repurpose the Lunar Module for LEO applications. While the company had, in the early 1980s, already begun transitioning into an electronics-focused subcontractor, Grumman Space Systems had not given up hope that the company could continue to produce some kind of space hardware. The Space Exploration Initiative was very warmly received in Bethpage and Calverton, where Grumman executives saw in it very clear applications for the company’s last flagship space technology programs: MPEM and the Space Nuclear Thermal Propulsion project. Alas, the shift of focus away from Mars and toward near-term operations in cislunar space doomed both of these efforts, as the need to validate the eCRV in flight and to conserve funding by down-selecting to the smallest number of new vehicles led Space Habitat’s planners to simply use one of the Block II eCRVs (which would be mounted to the habitat anyway as a semi-permanent lifeboat) as the transfer craft.

The European Space Agency was also open to cooperation with NASA on the development of the new, full-time Space Habitat, particularly after the agency clarified that the Space Hab was not intended to draw resources away from scientific use of Spacelab (indeed, that it could co-orbit with Spacelab so that routine resupply and crew rotation runs could be performed to each station on one mission). As that Space Habitat took clearer form, however, it became clear that ESA would be the junior partner in that particular vehicle. A decade of Spacelab operations had shown that microgravity was still a trouble spot for long-term space missions. Repeated attempts to mitigate the problem with pharmaceuticals and exercise had shown some improvement, but bone loss and muscle atrophy remained unresolved by the early 1990s (mechanical resistance suits, tested by the Russians on Salyut and Mir, were proposed in the West but had never been test-flown in the American program). As a result, particularly after the success of Wetlab in 1987, American space station planners had come to favor a space station with artificial, tumble-induced “gravity,” which would not only mitigate astronaut health problems but also simplify the design of life-support systems. Since all the experience with artificial gravity was concentrated in the US, it became clear early in the development program for the SEI Space Habitat that NASA, specifically Johnson Space Center and prime contractor McDonnell Douglas, would take the lead in planning the Space Hab.

For their part, the ESA was willing to collaborate on such a space station, offering an ERNO-built Command and Service Module to serve as a center of communications, thermal control, and power for the station while the United States supplied the S-IVC Wet Workshop and the Docking Module. While some of the proposed components could be repurposed from Spacelab spares and structural test articles, the unique, tumbling operation of the Space Habitat would require some new hardware designs, particularly in the solar arrays, whose ability to track the sun was compromised by the vehicle’s rapid movement. The Space Habitat CSM would ultimately become a collaborative effort between Entwicklungsring Nord and American aerospace giant Boeing, with questionable cost savings, but clear political benefits.

NASA’s next-largest international partner agency, the Japanese NASDA, did not bring as much space flight experience to the table as Europe did. Primarily concerned with establishing Japanese launch and earth-observation satellite capability, the agency had not paid much attention to human spaceflight or spaceflight beyond geostationary orbit until the late 1980s, when the surging Japanese economy permitted greater investment in such projects. When it finally did invest more, it produced work on-par with the best engineering in the US and Europe, notably the LE-5 and -7 rocket engines, which were by 1991 slated to power the “Sierra” reusable launch vehicle under development by Trans-Pacific Launch Industries. It was through TPLI that NASA sought to involve Japan in the Space Exploration Initiative, by arranging propellant resupply for the Lunar Transfer Vehicle and logistical support for Spacelab and the Space Habitat in exchange for the presence of Japanese astronauts on lunar missions.

By 1991, “Sierra”’s design had been frozen and work had begun on actual development. The original two-stage reusable design was expanded so that, on geostationary missions, payloads would be pushed into their transfer orbit by a Castor 30 stage, a cheap, expendable solid rocket derived from Thiokol’s Minuteman missile. With the end of the Cold War, Thiokol found its Minuteman production lines underutilized, and so was eager to sell quarter-length derivatives to TPLI. As part of the contract, Thiokol also signed an agreement with Ishikawajima Heavy Industries of Japan to manufacture a limited number of Castor 30s on license, for Japanese government payloads.

While Sierra’s first stage was a fairly narrowly-defined VTVL rocket stage, the second stage was, upon further analysis of the Low Earth Orbit market, redesigned into a vehicle that blended the capabilities of a dedicated spacecraft and a launch vehicle stage. Unlike airplanes, rockets, even partially-reusable ones, had no clear way to salvage a payload in the event of a loss-of-mission. The loss of Magellan had underlined this shortcoming, leading TPLI to search for a way to provide its customers with a way to ensure that their payload, whatever happened, could at least be brought back for a second launch attempt. The most effective way to provide this capability, Martin-Marietta’s engineers found, was to provide a payload bay, nestled between the second stage’s two propellant tanks. The bay, pressurized during ascent for greater structural integrity, would vent its residual atmosphere out into space before opening a clamshell door, exposing the payload and any attached booster to the vacuum of space. In the event of a first-stage failure, the second stage could separate and recover the payload, returning it safely to Earth. Furthermore, the payload bay could also be occupied with a reusable pressurized cabin, allowing the second stage to provide crew rotation and unmanned logistics to the growing fleet of space stations in Low Earth Orbit.

Pleased at the potential development of yet another partially-American reusable launch system, this one aiming to be fully reusable and optimized for payloads which were individually undersized for cost-effective launch on the Space Lifter, the NASA Launch Contracting Office signed a Memorandum of Understanding with TPLI. In the document, TPLI committed that they would aim to design their vehicle to meet NASA’s stringent requirements of safety and reliability for institutional launches and potentially even the higher standards for manned flights, and that should this condition be met, NASA's LCO would help work with the rest of the agency and other government organizations like the FAA and USAF to see the Sierra launcher certified to be a competitive bidder for the launch contracts that the office was responsible for negotiating and approving. A particular eye was given to using Sierra for propellant transfer to the Lunar Transfer Vehicle for the SEI. This memorandum was signed in a visit by the office’s Director to TPLI’s office in Boulder in which he met with the heads of the joint venture and senior members of the design team. It was viewed by NASA as a gesture of support for a new, partially-American firm helping further the development of reusable vehicles, without spending a dime of government money that wouldn’t have been spent otherwise--a point on the mind of an Office whose Director had quite recently had to repeatedly defend the Agency’s near-sole-source relationship with STC for launches of all large American payloads.

Whereas NASA’s agreement with the ESA on Space Habitat and reentry capsule development had been met with fairly muted protests, NASA’s agreement with TPLI (and through them Mitsubishi and NASDA) was decried by STC as a subsidy to foreign interests almost before the ink of Cizadlo's signature was dry. Pointing out that TPLI had yet to even reach Low Earth Orbit, STC officials aggressively lobbied their counterparts at NASA to concentrate on using the Space Lifter, with its long flight history and considerably greater per-launch payload, to support the Space Exploration Initiative and other institutional launches. While an understandable reaction in a purely mercenary sense (it is natural, after all, for a company with a de facto monopoly to attempt to retain that monopoly), this initial panicked reaction annoyed many at NASA, and led senior NASA officials (including Dan Goldin) to double-down on their commitment to the partnership with NASDA and TPLI. At no point did NASA indicate that TPLI would be receiving a monopoly on SEI launches (indeed, the limited payload of the Sierra launch vehicle compared to the Space Lifter made the latter necessary to realize the program’s goals). As such, STC’s attempt to lock TPLI out of even a partial share came off as presumptuous in the extreme. Furthermore, the partnership with NASDA and TPLI held the promise of a true redundant alternative to the Space Lifter, which could operate at a lower cost than the remaining Titan IIIs in the event of a future Space Lifter failure, and whose availability was not at all in doubt (unlike the USAF Titan IIIs, which were assigned to national security payloads first and foremost). The Space Lifter’s year-long stand-down had thrown NASA’s dependence on the launcher into sharp relief, and no one wanted to risk a year-long stand-down during, for example, the fueling campaign for an interplanetary craft on a tight launch window.

The memorandum, and the controversy surrounding it, was a boon to the joint owners of Trans-Pacific Launch Industries, which had begun to suffer serious financial strain as a result of the long Sierra development process, the bursting of the Japanese asset bubble, and the decline of its missile business. Martin had been relying on Mitsubishi Heavy Industries (and in turn the Japanese government) to supply a great deal of the capital to finish research, development, and testing of the Sierra launch vehicle, which had been planned for a late 1995 or early 1996 introduction to service. The collapse of the Japanese asset price bubble in 1991 severely hampered Mitsubishi’s ability to pick up Martin’s slack, and for a time Trans-Pacific Launch Industries appeared to be on the verge of failure. The resources of the Japanese portion of the partnership were newly constrained just as the company was beginning to to see the expected ramp up in expenses for the preparation of flight test hardware, component testing, and the conversion of launch sites at Tanegashima and Cape Canaveral.

The memorandum of understanding was a critical sign to commercial launch interests that NASA was as interested as NASDA in seeing the Sierra fly, and its approval was cited as a cause in several high-profile launch contracts TPLI secured in 1993, such as the Iridium and OrbComm low-Earth communications constellations. The down payments on these contracts went almost immediately into the funding of the beginnings of the test programs. The LE-7 was in its final lifespan risk-reduction test stand firings to prove it could reliably serve the role which Martin had demonstrated with the lower performance RL-10 on their Terminal Descent Demonstrator over the past five years. At the same time, with the initial proving tests completed, the first LE-7 production units had already been shipped for integration. Work was underway with the first boilerplate Sierra stage at Mitsubishi’s Nagoya Aerospace Systems Works in Tobishima and Martin’s first Fuji upper stage ahead of the testing of the systems at Mitsubishi’s Tashiro test site and Martin’s old TDD launch site at White Sands Missile Range.

Even as Mitsubishi began their testing of the Sierra first stage, TPLI launch teams were working out the infrastructure that would support Sierra’s flights to orbit. Three launch sites were identified for the the rocket to serve the Japanese market and the American market for commercial LEO and polar orbits. Japan’s site would consist of a new launch complex constructed at the Tanegashima Space Center which already launched their American-derived N-1, N-2, and H-1 rockets. The American sites would be more cumbersome, as Martin proposed to have TPLI adapt their existing infrastructure at Vandenberg and Canaveral. This would save on development costs, but would require working around the requirements to maintain a mothball capacity to launch Titan III rockets until TPLI could convince the Department of Defense that Sierra could meet their needs to back up Lifter. Thus, for the moment only two cells of the Titan Vertical Integration Building at Cape Canaveral and one of the two Titan launch pads (LC-40) would be converted. A similar Solomonic distribution was happening to the facilities at Vandenberg, with SLC-4 West to be converted for Sierra while for the moment its twin at SLC-4 East would remain on call for Titan. As crews set to work pouring concrete for new launch sites and maintenance hangars in two countries and designing the transport infrastructure to ship TPLI’s stages across half the globe, the first firing of an integrated Sierra first stage took place late in 1993 at Mitsubishi's Tashiro Test Facility in Ōdate. While the tests of the stage went well, the LE-7 suffered with issues taming its staged combustion cycle to the level required for rapid reusability. Still, the first “battleship” Sierra stage was deemed ready for the 1994 vertical launch-and-landing flights from a launch site at Tanegashima space center. The Fuji stage was proving more of a challenge, as Martin worked through the issues of making a space worthy orbital vehicle fit within the mass margins of a flight-worthy upper stage. It wasn’t until late in 1994 that the first completed vehicle was integrated and ready to be shipped to the White Sands Missile Test range. However, with the financial side of the business temporarily secured, TPLI was still progressing close to schedule.

The last major international partner NASA courted in the 1990s was Roskosmos, the successor agency to the Soviet Union’s civil space program. Strapped for cash, the agency was eager to leverage its technological expertise for hard currency. While the Americans and Europeans had developed man-tended, short-term space stations, the Soviets had, since the 1970s, been experimenting with ever longer manned stays aboard the Salyuts and then Mir. As the US pivoted from a short-term, LEO-focused program to longer-term missions further out, it was this expertise that NASA coveted. While there were questions as to how much Soviet experience could contribute to the notional partial-gravity laboratory in LEO, there was no question that the hardware the Soviets had developed for Salyut and Mir would prove useful on the planned Lunar Orbit Space Station, and on the in-development Habitat Module for Spacelab. Under the terms of an agreement signed between NASA, Roskosmos, and ESA in 1992, the Russian space agency would provide logistical support and life-support hardware for the Spacelab habitat module, and provide assistance in the development of the life-support system for the notional Lunar Orbit Space Station. For their part, the Russians were happy to have a contract for more flights of the Berkut space shuttles, and a contract to develop propellant-transfer capability for the Groza upper stage so that it could serve as a backup in the event of Lifter or TPLI failures. Such contracts were merely one of several ways the Russian program was seeking to secure the funding necessary to navigate successfully through simultaneous financial, technical, and social crises.

The dire financial straits in which the Soviet (and then Russian) space program found itself in the early 1990s opened the way to new ventures that were, in some respects, more exotic than logistics for the lunar program. In a (somewhat desperate) effort to find alternative sources of income for the Soviet manned space program, the Soviets had, as early as 1987, begun reaching out to laboratories in non-Warsaw Pact countries and offering them access to laboratory space on Mir at a price competitive with Spacelab. The several-month stand-down of the Space Transportation System in the aftermath of Magellan’s loss gave the Soviets a temporary advantage, which they used to the greatest possible extent, launching experiments designed not just in neutral countries like Brazil, India, and China, but also American allies like Japan and West Germany (and with even US firms expressing interest before Lifter’s return-to-flight restored the earlier status quo). More important than the experiments themselves were the diplomatic and commercial contacts that Russian managers made outside the Warsaw Pact, in particular, with Jeffrey Manber, formerly of the Office of Space Commerce (of the US Department of Commerce), and telecommunications entrepreneur Walter Anderson.

As the 1980s gave way to the 1990s, the Soviet Union’s economic and political structures continued to erode at an ever-faster rate. Resource shortages were rampant, and the new, more-open government was not having apparent success in combating them. Worse, ethnic separatist movements had begun to erupt from one end of the USSR to the other, from the Baltic states (whose citizens had fonder memories of their brief interwar independence than of their occupation by the Soviet Union) to the Muslim-majority SSRs of central Asia. Matters came to a boil in 1991, when an attempt by the KGB and Communist Party hardliners to restore order and roll back Gorbachev’s reforms backfired drastically. On December 26, 1991, the Soviet Politburo merely recognized reality when it dissolved itself and handed control of the Red Army (and other branches of the Soviet military) to what was once the Russian SFSR, now simply the Russian Federation.

The end of the Soviet Union did not spell the end of their manned space program, though it did severely complicate it. For one, the primary assembly plant for Russia’s main satellite launcher was now in a totally different country, and under the ownership of an entirely different government. For another, the Baikonur Cosmodrome, from which the majority of Soviet satellites had launched, was in still another country. Finally, the Russian space program was in dire financial straits. Not only was it now to be maintained by a considerably smaller economy (the Russian Federation’s per-capita GDP was roughly ⅓ the Soviet Union’s pre-dissolution per-capita GDP), but it had to fight for its funding in the court of public opinion. In the face of breadlines, drug addiction, an explosion in organized crime, and mass unemployment, appeals to the heroic legacy of the Soviet Union and Russia’s place as a world power fell, more often than not, on deaf ears.

Like two satellites separating from one another after orbital insertion, the Ukrainian and Russian space programs began to diverge very shortly after Ukraine gained her independence. Russian military officers very quickly came to regret the decisions of their Soviet predecessors to terminate Soyuz and Proton production. Though the anemic Russian economy did not provide enough funding to launch many satellites, enough had to be manufactured and launched (to complete the GLONASS navigational system, to replace aging Molniya-orbit communications and warning satellites, and to perform other critical military and civil infrastructure tasks) that they would swiftly have exhausted the remaining supply of warehoused Soyuzes and Protons. Ironically, what saved the Russian program from total dependence on Ukraine was Raskat’s reusability--the 10 boosters in the Russian fleet could, assuming no losses, supply the Russian Federation for most of the 1990s and even beyond (depending on maintenance quality). The Groza core stage was a more troubling matter, as it was expendable, but Ukraine was dependent enough on Russia for imports and exports that purchases were, for the moment, reliable. It was a difficult dependence to swallow for a military hierarchy so used to autarky, but there was no way around it--restarting Soyuz or Proton would take at least several years and a larger investment than the Russian Duma was willing to make in 1992.

NASA’s partnerships with ESA, NASDA, and Roskosmos were critical to the effort to shepherd the Space Exploration Initiative through the US congress. Though there was no Congressman or Senator with a particular grudge against NASA in the 1990s, American legislators were eager to cut off small chunks of the NASA budget in order to reduce taxes or at least to shore up their own “fiscal conservative” credentials. While there was no serious effort in the discussions over the FY1992 and FY1993 budgets to utterly defund SEI, there were proposals for reductions in funding and extensions to the schedule. A memo circulated at NASA headquarters in early 1992 contained a copy of the infamous “fusion never” graph, which, when first published in 1976, stated that, if the 1978 level of fusion energy research remained constant, there would never be a commercial fusion reactor in the US. As the memo elaborated, “much as a rocket can burn a lot of fuel and stay on the launch pad if its weight is greater than its thrust, a program can, in fact, burn a lot of money without doing anything.” NASA staffers and industry lobbyists set to work persuading congressional staffers of the importance of funding the now-international Space Exploration Initiative, implying dire consequences would result from a lack of American leadership on an international program in space.

The effort was much harder than it might have been a decade or two earlier, as, without the Soviet menace, it was difficult to paint a plausible scenario of any nation other than the US taking the lead in space. Legions of staffers worked long hours to find new talking points. Eventually, they hit on a number of points that were emphasized to varying degrees depending on the relevant Congressman’s interests. The collaboration with the Russian Federation on the Space Habitat partial-gravity laboratory appealed to both security-interested congressmen anxious to prevent a drain of Russian talent into Iran, Iraq, or China, and to optimists eager to create in Russia a sincere American partner. Collaboration with the European Space Agency and Japan similarly appealed to those who recognized the long-term importance of America’s allies in supporting future interventions around the world. The possible utility of the the new Space Habitat as a laboratory for biomedical research with terrestrial applications was also brought up, though the similarity of that argument to old promises of microgravity wonder-drugs was not lost on veteran Congressmen who had heard those promises in the 1970s. Together with the typical “jobs at home” arguments that easily mobilized space state congressmen, these arguments helped keep SEI’s various programs funded even as the Bush Administration gave way to the Clinton Administration.

The success of the Space Exploration Initiative must be contrasted against the failure of its counterpart in the field of particle physics, the Superconducting Super Collider. As it had treated spaceflight, the US spent much of the Cold War treating fundamental physics research as a field of soft power projection, a way to demonstrate American supremacy in science. Officially begun in the mid-1980s, the Superconducting Super Collider was to be the latest in a line of increasingly large and sophisticated colliders built in the US. However, it floundered on the rocks which NASA and its contractors had narrowly avoided during SEI and during the earlier effort to fund the Space Transportation System. Like the NASA programs, the SSC lost a great deal of support in the aftermath of the Soviet Union’s collapse, as the Russian Federation was in no place to challenge America’s leadership in particle physics, and as budget-cutters, led by Kansas congressman Jim Slattery (who, at the same time, was also challenging the B-2 stealth bomber) eyed the program as a source of Peace Dividend savings. Unlike its aerospace counterparts, its supporters were concentrated almost exclusively in Texas, where it was to be built, and to a lesser extent at the handful of research laboratories supporting the effort. This limited greatly the amount of Congressional support it could rally. The SSC also failed to generate serious partnerships with American allies in Europe and Japan, who, for their part, preferred to develop their own experiments, whether under the auspices of CERN or under their own national and university laboratory systems. Most crucially, the SSC’s budget had inflated sharply since the project’s initiation, from $4 billion quoted in 1987 to $12 billion in 1993. While this was still a drop in the bucket compared even to NASA’s cost estimates for SEI, the sudden budget increase embarrassed the program’s backers and opened them up to allegations of mismanagement and incompetence, particularly since a great deal of the program’s staff were new to particle accelerator operations. SEI’s planners, by contrast, were able to point to the low cost of Space Lifter operations over the years and the fact that that program had come in on-budget (after adjustment for inflation) during its development to defend their programs from similar allegations. Furthermore, the comparatively low costs of fixed-price launch contracts (which were to be used for propellant launch to Low Earth Orbit) appeared to ensure that, when NASA quoted a $40 billion program cost, it would come in at that value.

As George Bush's first term proceeded past his first midterm elections, his space troika had proved successful in creating a plan and marshaling support in NASA's management, on Capitol Hill, and around the world for the Space Exploration Initiative. While they had achieved a major turnaround in the fortunes of an agency that had seemed rudderless after the Magellan failure, the task remained to see the funding that had been secured against the headwinds of the Peace Dividend turned into real hardware. With the approval in the FY 1992 budget in late 1991, the first funds were made available to power the program out of the glide it had been in since the Return to Flight and turn the early studies laying out the Space Exploration Initiative into measurable progress and real hardware.
 
Last edited:
However, with the financial side of the business temporarily secured, TPLI was still progressing close to schedule.
Ok, so I'm having trouble visualizing this. What I've gathered is that the Sierra uses a Falcon 9 burnback system for first stage reusability, once-around and parachute splashdown for the second stage, and puts the payload in the intertank (which would massively cut into payload, because there's no way to have a payload diameter greater than the rocket, or a payload height greater than the intertank) and uses OTL LE-7 and LE-5 engines, with no option for SRBs. Is this correct?
 
Ok, so I'm having trouble visualizing this. What I've gathered is that the Sierra uses a Falcon 9 burnback system for first stage reusability, once-around and parachute splashdown for the second stage, and puts the payload in the intertank (which would massively cut into payload, because there's no way to have a payload diameter greater than the rocket, or a payload height greater than the intertank) and uses OTL LE-7 and LE-5 engines, with no option for SRBs. Is this correct?

Basically, though the payload size restrictions aren't all that severe--between the fact that Sierra uses LH2 and the fact that, like all reusable vehicles, it's oversized compared to an expendable vehicle, the intertank on the second stage is actually quite voluminous--the diameter is over six meters. As to height, the intertank is longer than one would expect on an ELV, as greater second-stage surface area also reduces the thermal loads on the stage during reentry. Lifter's long domination of the market has also led satellite builders to optimize for its own 6.6-meter fairing diameter--that is, satellites ITTL are, in general, a bit squatter and fatter than IOTL to allow for easier multiple-payload launches aboard Lifter. The Sierra's payload bay is big enough to take most commsats ITTL, though only one at a time.
 
To augment what Polish Eagle said, here's a sketch of TPLI's vehicle using the output of one of our spreadsheets. Engine bays and such aren't included, but I think you can figure out where they'd go. The payload envelope is just over 5.5m diameter and just over 7.75m long--the same diameter allowed for payloads inside the lower bay of the Space Lifter Multiple Launch Adapter. The second stage doesn't necessarily only stay up one orbit--for satellite deployments to LEO or GTO it'd stay up multiple orbits to phase back to land at its launch site, and for possible propellant or station logistics it'd stay up long enough for docking and offloading its supplies.

upload_2017-10-8_14-13-33.png
 
Last edited:
...for Constitution remained, aerodynamically, a brick. She had never been optimized for slow, level flight, but for the job of lifting a payload to the upper fringes of the atmosphere. As her speed dropped, so did the lifting force on her wings, and she acted more and more like the wingless fuel tanks she counted among her ancestors. Unassisted, she would not have managed to return to land, and would have flopped sadly into the Atlantic. This had been obvious to her builders, and they had given her a way to overcome that limitation…
That would be the jet engines I trust.


....CRV capsules.

Through the use of more modern manufacturing techniques and lightweight alloys, and the use of a very minimalistic “service module” (essentially a set of retrorockets with explosive separation bolts), the entire spacecraft came in at under 4,500 kilograms (a further 1,000 kilograms was allocated for a maneuvering bus that, launched with the CRV on an Ariane 2 rocket, would guide it to a docking with Spacelab). It was this design that was frozen in 1985 as “Asclepius,” after the Greek god of medicine.

By 1989, the European CRV was far-enough along that almost every analysis conducted at NASA’s Johnson Space Center found it to be the fastest and most cost-effective path to a lunar return vehicle. While the Space Shuttle was well suited to returning crews from Earth orbit, the higher speeds of lunar entry would subject it to far more heat than it could handle, and the added weight of the Shuttle's cargo bay and winglets would be of minimal use on circumlunar or lunar orbital flights. The European capsule, by contrast, was lighter to begin with, and its simpler shape made uprating its thermal protection system a smaller challenge. Under ESA’s proposal, an enhanced, longer-lifetime, more-durable variant of the Spacelab Asclepius CRV would serve as a crew capsule that would carry the crew to Low Lunar Orbit and back, propelled by a reusable American Lunar Transfer Vehicle. The capsule, with room for 4 astronauts, would separate from the LTV as both spacecraft approached Earth, allowing the LTS to propulsively brake into orbit for reuse. This plan, which European engineers dubbed “Hermes,” achieved a number of goals for both NASA and the ESA....
Hmm.
Asclepius is being pushed pretty hard, to extend a craft designed for 4 people for hours, into one that will host them a week or more. They only need it in transit to and from the Moon, assuming all 4 descend together, but considering it is 3/4 the mass of Apollo CM and yet has an extra person aboard. I know the CM is considered by some to have had more space than needed and that for low orbit missions shuttling to a station, as many as 5 astronauts could have been fit in Apollo CM, but it seems risky to me. Advancing tech, with Asclepius designed in the later 80s versus CM in the early 60's, might be a big help, but it gives me some pause.
...Ultimately, many more Hermes than Asclepius capsules would be built, as only the latter was planned for actual operational use on either new program.
Shouldn't that be "former?" The old plan was the original Asclepius for everything but the moon mission, but then Hermes was chosen for interorbital transfer as well, and on the "one craft for all" principle of avoiding multiple types when one can stretch for all, Hermes becomes the lifeboat too? They wind up making many Hermes but only prototyped Asclepius?
Though the transition from “permanent Spacelab” to SEI would eventually consign Asclepius to that status of a technology development/prototyping program (one prototype was completed, but never meant for flight), the work done on the Block I eCRV massively accelerated the Block II spacecraft’s progress once that program finally received the green light in 1991.
Like that?
....
While Sierra’s first stage was a fairly narrowly-defined VTVL rocket stage, the second stage was, upon further analysis of the Low Earth Orbit market, redesigned into a vehicle that blended the capabilities of a dedicated spacecraft and a launch vehicle stage. Unlike airplanes, rockets, even partially-reusable ones, had no clear way to salvage a payload in the event of a loss-of-mission. The loss of Magellan had underlined this shortcoming, leading TPLI to search for a way to provide its customers with a way to ensure that their payload, whatever happened, could at least be brought back for a second launch attempt.
Um, put all payload on top, as per Lifter, and have a Launch Escape system of the kind that even ITTL had been installed on Mercury, Apollo, and Soyuz, as well as provided via supersized hypergolic emergency engines for the Lifter-Shuttle?

It may be argued per below that this is old fashioned and obsolete because of better ways, but it hardly is unclear how to do it. Regardless of whether the second stage is designed to return to Earth in reusable form or not, it certainly would be possible to design Sierra analogously with Lifter in this respect, and give every payload an escape system not associated with the second stage. To the objection that this is wasteful since the thrust is only ever used in emergency--well, it works in the Lifter Orbiter to take the dead weight penalty of the surplus engines, so we are told. And considering Lifter suggests maybe we don't need to replicate the capabilities of Apollo LES, with its 12 G acceleration. I don't think the ATL Orbiter can reach that degree of separation speed and distance--and if it is instead toughened to take a powerful blast coming from behind, that gives the rear end a hard kick which itself augments accelerating it further away, we might not need as much, if it is rugged enough to remain intact, sealed and controllable after the blast. This suggests a tough payload bus base, combined with adequate thrust to pull the capsule briskly enough to bring the blast impact down to acceptable levels would save payloads as well as astronauts.
The most effective way to provide this capability, Martin-Marietta’s engineers found, was to provide a payload bay, nestled between the second stage’s two propellant tanks. The bay, pressurized during ascent for greater structural integrity, would vent its residual atmosphere out into space before opening a clamshell door, exposing the payload and any attached booster to the vacuum of space. In the event of a first-stage failure, the second stage could separate and recover the payload, returning it safely to Earth. Furthermore, the payload bay could also be occupied with a reusable pressurized cabin, allowing the second stage to provide crew rotation and unmanned logistics to the growing fleet of space stations in Low Earth Orbit.
Really? "Most effective?" I suppose first stage failures include relatively gentle ones, such as an engine going out in the middle of the burn, and then firing the second stage immediately could indeed achieve separation--but only gradually, not at the 12 G's or so Apollo LES accomplished for the CM or even the half that or less the TTL Lifter-Orbiter's emergency engines achieve. It can only separate at design separation thrust, which typically is in the ballpark of 1 G or less--e of pi has given a rule of thumb of about 3/4 G. Unlike Lifter Orbiter, it would be prohibitively costly in several ways to give the entire second stage multiple G's emergency thrust capacity--not only would the engines themselves be dead weight but designing the stage to bear more than routine maximum G's would add weight needed only in this emergency.

Now I suppose maybe, if this approach were taken, routine maximum G's are pretty high, given by thrust on the nearly dry first stage at burnout. Maximum G's just prior to second stage burnout is different; that is at much lower thrust with the second stage tanks nearly dry--but if it can take say 3-5 G's with first stage burnout, it can take them under emergency thrust. In that case, if it were possible to have emergency hydrogen burning engines that could manage 4-8 times normal second stage ignition thrust, their considerable added dry weight would be offset by making both stages bigger and routine operational expenses moderately higher, a tradeoff of the goal of cheaper reusable operations versus achieving credible emergency safety in this contingency. But it would be a substantial tradeoff I think.

Meanwhile it is not clearly better in the sense of being more effective than a more traditional capsule on top approach with dedicated emergency engine saving just the payload capsule itself--achievable acceleration will be lower to begin with.

More important the idea of crew separation systems in the 60s were mainly aimed at covering the possibility that stage failure will not be gentle; it may involve one stage exploding quite rapidly and suddenly, with little to no warning. If the first stage blows, at any time before or after ignition, it seems highly unlikely that the second stage can remove itself from the explosion fast enough to avoid levels of damage that surely must compromise all its functions. Continuing to fire its engines after taking blast forces centered just tens of meters away risks starting a second explosion within the second stage, if indeed the blast is not sufficient to trigger that directly and immediately. Even if it can fire without such a result, all the structure touted as protecting the payload or even God help us crew is under question, how well it can survive an entry even drained of propellant will be quite unknown. Of course if the primary first stage blast or shrapnel damages the second stage we are then quite likely to have a secondary explosion, with precious cargo live or otherwise nestled right in the midst of the propellant tanks; it seems like taking shelter from machine gun assault by ducking behind crates filled with dynamite to me!

Traditional capsule escape from the top of the stack cannot guarantee survival in all circumstances, but it comes a lot closer than this, which at best would work only in certain limited modes of failure. Mercury/Apollo versions would work robustly in more circumstances.

To the remark that escape in a few circumstances is better than none, I doubt that if a first stage had a "gentle" emergency, that it would stay that way once the second stage fired. With hundreds of tons equivalent thrust from hot hydrogen-oxygen exhaust playing over the nose end of the first stage, and the upper stage pulling away at 10 m/sec^2 or so, this blowtorch has considerable time at considerable intensity.

I suppose Apollo LES was not primarily designed for separating from a failing upper stage, but it would have some chance; toughened bottom face might allow blast to bat it before rocket thrust could move it far; if a secondary first stage explosion were triggered the escape engines are already under way. Primary design might have covered a point-blank S-IV explosion as well as a larger but more distant S-IC explosion, perhaps neither could be survivable without prior separation under thrust, or possibly upgrades, in the form of weight-costly blast shielding (which might be cleverly integrated with mounting or thrust structure to be sure to lower the weight penalty) might judo deadly blast into separation assistance at least in some ranges of likely blast, and blast shielding might also protect against shrapnel. Anyway, it is a chance of survival; the Sierra plan provides zero protection against upper stage mishaps, and as I say proximity to a first stage blast, with the attempt to separate seeming quite likely to trigger one, seems likely to damage the integrity of the second stage anyway.

To argue that the tradeoffs favor the Sierra plan, versus the alternative of putting the payload on top with option of removal and possible blast protection measures, have some advantages to set against the plain drawbacks is reasonable, if one can show that the contingencies it does protect against are on the whole more likely to occur, without the attempt at early firing the second stage raising the probability of a nearby explosion to near certainty as intuition suggests it would. Such tradeoff arguments would involve the assertion that the second stage will be far less likely to suffer a failure of any kind--for any kind of failure, even one not causing detonation, does leave the valued payload embedded in it with no further escape option than its successful landing, and probably with unspent hydrogen and oxygen propellant surrounding it, which greatly raises its descent weight and thus defeats normal braking and landing expectations, while posing an ongoing hazard; some means of reliably venting it all would need to be provided. All of the possible second stage failures must be set at naught, or such low probability that losing the ability to escape the whole stack top-mounted payloads could have is a small risk versus advantages. Then since I doubt it can be proven the obvious risk of nearby blast from a failing first stage is reliably survivable, it has to be shown that these too, which top-mounted capsules can more likely escape, are so rare as to be risked with little predicted probability. And the positive advantages shown to pay off so handsomely that they outweigh the accepted greater risk with considerable advantage.

And so I ask, how advantageous is putting the payload inside the tank, really? There on the way up they are protected against air drag and so on, but the same is true of a top loaded payload under a shroud. But the dead weight of a shroud can be ejected; the tank structure, which, it is explained, is stretched and elaborated thus admittedly adding weight, stays all through the launch. That makes it more reusable, with no costs associated with disposable shroud. Shroud ejection is a risky staging event, as Skylab's fate proved, and it is avoided. Trading off a bit more massive stage set, since we do not eliminate shroud weight, is arguably a win therefore. A top-mounted system might work the same though; an openable shroud that a launch escape system removes with it can be closed and recovered for reuse as an integral part of the stage after all.

In addition to serving as an alternate shroud going up, the intertank bay is touted as a downmass haven as well. This is something that would indeed be dubious to do on the tip; a reusable integral shroud might be made of titanium or some such to endure entry heat, but making it able to protect fragile downmass seems much harder. Nor would it be easy to enable such a a design to accept variable down mass, due to obvious center of mass issues; cleverness in the tank placement and thus payload bay location could put downmass at the center of mass so any amount of it leaves dynamic balance unchanged. I note that the description mentions air seeping out as the rocket ascends--implying that as it descends, air will seep back in and since during early entry this will be red-hot plasma, presumably there is some means of maintaining balance with presumably much cooler gas somehow preempting this incursion of damagingly hot gas maintaining no pressure difference?

We might presume that we can rely on the spent stage doing its job of protecting its own structure as well as whatever cargo lies in the bay. This downmass capability, especially I suppose for human crews, is the real draw of the design. You just can't do this with a capsule on the nose design.

What you could and for cargoes needing to land again must do is include a full function return capsule in the rated payload--such an inclusion might allow elimination of protective shrouds on the way up.

But vice versa, how often will Sierra upper stages actually bring down downmass?

The design commits every launch to tradeoffs for a safety feature (or so it has been presented first) that seems overall somewhat dubious in its range of coverage, providing no protection from part of the predictable range of mishaps that a tip-payload removal system could more comprehensively cover; this is at the price of introducing weight in the form of the expansion of the intertank into a payload bay, which seems likely to introduce extra mass traded off against traditional payload shrouds. These might look heavy compared to the introduced bay extension mass, but they are dropped early in burns and so their mass does not count in full against payload, while the fixed payload bay does, ounce for ounce, due to being boosted all the way to orbit. It saves a staging event to be sure, though obviously failure of the payload doors to operate correctly in orbit would prevent deployment of a payload--which could then ride down to Earth again for another try. If the doors do open but then do not close correctly, the stage is liable to be lost, or suffer damage taking it out of service at least. This may be improbable but shroud separation also usually goes well enough, so neither is clearly superior in that respect. Every payload is as well protected as the to-me dubious reliability of depending on the limited acceleration of the second stage protects any load; with a tip mounted payload, customers could opt as they wish for higher payload and risking loss, which might be insured, versus incorporating their payload into a launch escape system similar to that used to raise the odds of astronaut survival. To be sure a tip load that can be removed by high acceleration escape system would be subjected to acceleration far beyond its normal design load and therefore such a system might not be useful for it--then again the thrusts of the removal rockets can be designed to be modest, at greater risk of not escaping blast in time--were they to rely on the proposed Sierra design instead they'd have no choice, not even be able to supply modestly greater thrust--5 G for a nominal 3 G max launch would provide better escape but probably not overstress something designed with decent safety factors for instance--not an option for payload bed in stage Sierra. The same event that destroys a tip payload because its owners elect to limit the thrust would with greater assurance wreck a Sierra upper stage, payload and all. Whereas an event a launch escape equipped system might survive if marginally would also wreck the Sierra for sure.

As launch mishap survival means then, it seems clear that far from being "most effective," this design is distinctly inferior as well as inflexible. It is only when we consider other aspects that it presents advantages and those of are questionable relevance to most launches. Even successful recovery is a dubious gain; launch has failed, and the payload must be placed in a different rocket to try again after some delay. If the entity launching can afford delay, they might do far better to simply insure the value of the payload and order another one, or have a spare on standby, to launch later, since later is after all when a saved one would go up. I think tip escape systems are clearly far more effective and also flexible, being tailorable to specific needs and completely omitted when insurance or simply risking loss will do. They may mass more when opted, but setting that against more comprehensive protection, I think the smarter thing to do is design Sierra for more capability, and focus on itself, leaving orbital operations to be handled by payload features while the Sierra launch control concentrates on recovering the second stage ASAP and independent of the orbital operations going on. With a simplified upper stage and no loitering around in orbit, using up maneuvering propellant to haul a mass irrelevant to orbital operations all over the sky, Sierra launches can be that much more simplified and thus reliable.
...Russian military officers very quickly came to regret the decisions of their Soviet predecessors to terminate Soyuz and Proton production. T
Here Soyuz refers to the launch rocket I believe, but isn't it also the case they have abandoned the Soyuz spacecraft? They have developed a kind of Shuttleski going up on the Raskat-Groza combination, analogous to Energia but with flyback boosters operational, I believe, but did they also develop TKS or is the spaceplane their only means of crewed access to orbit? I forget.
 
To augment what Polish Eagle said, here's a sketch of TPLI's vehicle using the output of one of our spreadsheets. Engine bays and such aren't included, but I think you can figure out where they'd go. The payload envelope is just over 5.5m diameter and just over 7.75m long--the same diameter allowed for payloads inside the lower bay of the Space Lifter Multiple Launch Adapter. The second stage doesn't necessarily only stay up one orbit--for satellite deployments to LEO or GTO it'd stay up multiple orbits to phase back to land at its launch site, and for possible propellant or station logistics it'd stay up long enough for docking and offloading its supplies.
From the sketch it seems that the VL flyback first stage is a squat cylinder, only slightly taller than it is wide. Given it too uses hydrogen for fuel it seems sized so the hydrogen tank is close to spherical--a dome-tipped cylinder as wide as it is tall, or a little shorter in height for the straight-sided part accounting for the dome ends, while the oxygen tank is a disk, nearly, of the same diameter but much shorter.

If it is actually ker-lox I suppose something similar applies with the fuel and oxygen tanks swapped, and given its relatively compact size I suspect maybe that is the case.

Then the upper stage is quite unusual for a developed used stage design though Japanese proposals for SSTO, and DC-X, have adopted it. Here though it looks like it might be a biconic shape. That would require TPS all over its surface, the cone part especially, because it would enter "side on." Or one could instead visualize it as a conical capsule entering with the circular base first. That is a bit problematic since the engines are there and if they are traditional flaring bell nozzles they would be subject to hypersonic air blast; they might be a plug nozzle arrangement but I think you'd have mentioned it. Entering biconic seems fairly likely, followed by gliding airborne at high speed (but subsonic) then flipping over to rocket brake to rest over a landing spot then hovering and lowering down.
 
@Shevek23

The latter remark--yeah, typo on my part. Corrected. Same with the LEs that @Dathi THorfinnsson pointed out.

The explosion risk isn't analyzed to be as great a threat as you suggest. Hydrogen and oxygen aren't hypergollic with one another, so a structural failure on the first-stage tanks would most likely not result in a deflagration or detonation, but just rapid venting--like Challenger or SpaceX's CRS-7. And the second stage, being an RLV itself, is rather more sturdily built than most rocket stages, so it could survive a bit of hypersonic buffeting (though things would get a bit dodgy around Max-Q). The Falcon 9 first stage on CRS-7 kept pushing through the wreckage of the second stage for a while IOTL before range safety put it out of its misery, as an example. TPLI's two parent companies both decide that it is better to try and work recoverability (even in the case of first-stage failure) into the second stage than to develop what amounts to a third spacecraft.

Downmass is actually fairly common--about a dozen or more times per year, between propellant servicing runs, station servicing, and satellite retrieval. They began cutting metal on Sierra in 1989, at which point NASA was getting pretty gung-ho about finally getting its Space Tug and had a station that needed servicing, so they had the idea that there would be a significant number of missions where they'd like to recover their docking apparatus. Furthermore, since Sierra was always going to be a LEO-optimized spacecraft, they had their own serviceable third stage in planning for a time. Add some ambition on Martin's part to sell KH-12 servicing flights to the USAF, and you've got a good reason for a functional payload bay.

As to venting, there are valves in the payload volume that can be closed for reentry.

As to your other points, the Sierra upper stage does indeed reenter side-on, and the Soviet spaceplane is their only way of putting crew into orbit.
 
The explosion risk isn't analyzed to be as great a threat as you suggest. Hydrogen and oxygen aren't hypergollic with one another, so a structural failure on the first-stage tanks would most likely not result in a deflagration or detonation, but just rapid venting--like Challenger or SpaceX's CRS-7.
I don't see what being hypergolic has to do with anything. Kerosene and oxygen are not mutually hypergolic either. Still, lots of ker-lox first stages, as recently as one year ago as I'm sure we all remember, go kaboom anyway.

I distinctly recall discussion in Kolyma's Shadow that hypergolic fueled rockets like the Titan or nixonshead's ATL R-6 Soviet launcher did not require their spacecraft to have rocket escape systems, just ejection seats, because actual hypergolic propellants do not detonate as sharply, due to being hypergolic and thus reacting on contact and not forming fuel/oxidant mixtures before a spark belatedly ignites them

"Hypergolic" means two substances react on contact, requiring no spark; combustable means you do require a spark to initiate the reaction--the thing is, in a failing active rocket stage, a spark is ready to be found. Even if the engines were to shut down, all kinds of things might provide that spark. Or might not--but that just gives oxygen and fuel to mix into a worse basis for detonation than ever!

As I recall, Challenger did blow up with a substantial detonation. Not surprising considering it was being blowtorched!

NASA did not install solid rockets capable of pulling the Apollo CM at 12 G's for fun. If trying to get out of range of a big explosion and its shrapnel were not in the cards, then they'd have trimmed that down to 3 or 4 G's if all they were worried about was giving time and height for a chute to deploy or getting far from a big fire. 12 G's was the compromise between killing the astronauts with acceleration versus killing them by leaving them too close to a blast wave source, and trying to outrun the speed of sound of a strong shock wave.

Is anyone here seriously saying LH2 and LOX kept in large quantities together in one structure pose no seriously probable threat of massive explosion?

The probability may be fairly large that if something major goes wrong warranting emergency escape, it still does not trigger non-hypergolic propellants to explode. Indeed in all the footage I've seen of launch rockets failing, sometimes there is no explosion, just a collapse.

But we are now talking about a design where the second stage, the one physically attached to the top of the first stage, ignites to push itself away. Now there is a hydrogen-oxygen blowtorch of great force and heat playing on top of the abandoned first stage. I think there is, under those conditions, considerably elevated chance of ignition, and worsening leaks to provide fuel to ignite, don't you?
...TPLI's two parent companies both decide that it is better to try and work recoverability (even in the case of first-stage failure) into the second stage than to develop what amounts to a third spacecraft.
Again, a rather weird word that is off the subject. Recoverability is not at issue. If stages do not malfunction this should work fine for recoverability. The issue here is surviving a major traumatic failure, involving heavy shocks.

Also, much as with Lifter, there seems to be a blind spot regarding the possibility of a failure of the first stage on the launching pad. This definitely does happen, but like the possibility that the failure will be in the second stage rather than first (and you just cited, as plausible and convincing evidence recoverable stages are more robust and can survive detonation of another stage, a Falcon 9 first stage flying through debris of a second stage that failed!) it seems the TL's engineers are blandly claiming that the probability is too remote to worry about.

Second stages, as a rule, have thrust that is initially below the weight of the fully fueled stage. This is useless for takeoff but works well for upper stages. Unfortunately that means that should it be necessary for the upper stage to escape a collapsing or burning first stage on the launch pad, it cannot do so. It would be necessary to install 4 times the thrust normally needed to stage, as contingency thrust to achieve a decent three G initial separation rate.

If the first stage goes pear-shaped on the pad, the second is not even "recoverable," let alone survivable.
Downmass is actually fairly common--about a dozen or more times per year, between propellant servicing runs, station servicing, and satellite retrieval. They began cutting metal on Sierra in 1989, at which point NASA was getting pretty gung-ho about finally getting its Space Tug and had a station that needed servicing, so they had the idea that there would be a significant number of missions where they'd like to recover their docking apparatus. Furthermore, since Sierra was always going to be a LEO-optimized spacecraft, they had their own serviceable third stage in planning for a time. Add some ambition on Martin's part to sell KH-12 servicing flights to the USAF, and you've got a good reason for a functional payload bay.
So, to be blunt here, very much as with OTL STS, a feature that designers really really want is incompatible with proper levels of launch safety--were human crews involved anyway. I note that you've never said humans will ride Sierra up to orbit, just that they might ride in its bay down--although if none ride up, that wastes the bay since the pressurized compartment will displace any other payload. Perhaps the compartment is brought to orbit another way, and installed into the bay in space? Bearing in mind that non-human cargoes are up for discussion here, not people, it may be well to take some risks, and look on the bright side of having taken steps to survive a portion of them. So they emphasize the positive, avoid mentioning the negative and hope no one picks up on it. They are not wrong if they are selling their system as one of improved recoverability for payloads, despite some mishaps, and the ability to return objects to Earth from orbit is such a positive as to be worth setting against acknowledged negatives.

But that is not what they say. They say their system is the most effective, and considering that the alternative method I have highlighted, being disparaged now as "designing a third stage," is ancient history and also more effective--effective for pad aborts, effective if the second stage fails, effective if any stage explodes--at the focused item of discussion, survival of stage failure, it is very misleading to say that when what they might honestly mean is "pretty good considering we don't want to compromise another feature we'd like to include we couldn't use with the most effective survival approach-also recovery inside the stage means we use the downmass feature in case of abort, so that's elegant!"

But that would be harder to fit on a banner ad, so why not go with a highly misleading and incorrect short phrase instead?
As to venting, there are valves in the payload volume that can be closed for reentry.
I considered that and dismissed it because that would be worse than letting hot plasma repressurize it. If the bay is sealed off it will come down with vacuum inside it, putting a collapsing pressure all around. Instead some other source of gas is needed, one that is not too hot, but also one that exists, to match the rising pressure outside. Perhaps if a cryogenic gas like liquid nitrogen is supplied, and can be mixed with suitable amounts of outside air to fill the bay at a low temperature, the slower it goes the cooler and denser the outside air, and eventually you can empty the LN2 tank because outside air is now cool enough.
 
I considered that and dismissed it because that would be worse than letting hot plasma repressurize it. If the bay is sealed off it will come down with vacuum inside it, putting a collapsing pressure all around. Instead some other source of gas is needed, one that is not too hot, but also one that exists, to match the rising pressure outside. Perhaps if a cryogenic gas like liquid nitrogen is supplied, and can be mixed with suitable amounts of outside air to fill the bay at a low temperature, the slower it goes the cooler and denser the outside air, and eventually you can empty the LN2 tank because outside air is now cool enough.
Or a compressed gas like nitrogen could be supplied with which the bay is pressurized on-orbit before descent after the doors are closed, and then the bay is kept sealed from thousand-plus-degree outside air on descent. Which is what they do. It takes a lot less gas than trying to dilute plasma down to something that won't melt your spacecraft from the inside.
 
Chapter 15: Spool-Up
“A marginal SSTO is a great TSTO.”
“By extension, a marginal TSTO is a great 3STO.”

Hand-written notes on NASA MSFC memo during STS planning, 1969​


Chapter 15: Spool Up

With a loud whine, hydraulic actuators opened the air intakes for Constitution’s twin Auxiliary Power Units. Supersonic air slammed into the compressors of the small jet turbines, beginning to turn their cores. As the lights on the cockpit displays indicated the doors locked open and the RPM gauges rose off the stops, Young and Crippen began the startup procedures. Crippen read off the checklist as Young opened the valves which fed peroxide, then kerosene to APUs. The decomposing peroxide, flashing to superheated steam as it passed the catalyst beds, hit the starter turbines on the two little jet turbines. As they drove the rotation rates above the critical threshold for start, the kerosene flow was set and a burst of spark lit the burners. As the turbines continued to spool on their own power, the peroxide flow was cut, its job done. The whine of the units, conducted through the frames, contributed a barely audible tone to the cabin as Constitution left her wake far behind.

With the two generators picking up the loads of
Constitution’s equipment, another gauge showed that the bleed pressure to the main engine bank was also rising to nominal. As Crippen cross-checked the gauges and called off the checklists, Young began opening the engine inlet doors, starting with the inboard pair. As with the smaller auxiliary units, ram compression on the fans was enough to get the core turning, but a burst of compressed air was necessary to boost the engines above starting speed. As they spun, the fuel pumps began injecting a fine mist of jet fuel into their combustion chambers. The rarefied but high-velocity inlet air might have been a problem, but Constitution shared her powerplants with the latest generation of fighters and bombers in the American fleet. Their original designers had known engines might have to light at such conditions, and as with most systems on the Lifter, the jets were over-engineered and redundant. She could have flown with only six active engines, and as it was, 7 out of her 8 lit on the first attempt. The last, moments later, spooled up under the full attention of the flight crew and controllers back at Houston examining the telemetry, even as their fellows monitored the ascent of Endeavour to orbit. The whine of the APUs was lost in the distant rumble of the engine bank.

The starters were secured, and Young adjusted the throttles for their flight back to Kennedy. Opening valves in the tanks, they began to bleed off the residuals left over from the launch, sloshing in gutters on the former sides, now the bottoms of her vast propellant tanks.
Constitution briefly acquired a tail of RP-1 and boiling LOX vapor to join the wide blended contrail from her jet bank as she completed her conversion from rocket to aircraft. The roar of the engines trailed far behind with the sonic shock-wave as Constitution started her Mach 2 speed run back to shore and Kennedy Space Center.

As the Space Exploration Initiative continued to spool up efforts to turn Bush’s selection of an Option B lunar access system into reality, new-start projects continued to sprout like mushrooms after rain in the fertile grounds of the increased budget. The most major projects, like the permanent space habitat and a reusable tug capable of accessing cislunar space, were key to the initiative. These large development projects were hotly debated within NASA and industry sources, as managers sought to define the program’s objectives and firms competed to shape the SEI’s version of space exploration hardware to fit the bids they believed they could put forward. The new surge in development money had other beneficiaries, though, as projects less closely related to the Space Exploration Initiative fought to tie their pet concepts to the new lunar focus in hopes of securing some fraction of the funding. The benefits for the projects which succeeded in securing such windfalls were large, but even securing a place on the starting line was only the beginning of the battle to bring a project to fruition. Managers would have to guide their projects through all the ramp-up challenges from contracting to delivery if the Space Exploration Initiative was to pull the American space program out of its post-Magellan doldrums and point it back on a course to the moon. The cost of failure was potentially high, but a success could ignite a new blaze of American innovation even hotter than that unleashed by the Apollo program.

In Low Earth Orbit, the Space Exploration Initiative was most visible in the new Space Habitat, which came to be known as the Armstrong Partial Gravity Laboratory. Named, officially, after Harry George Armstrong (the founder of the USAF’s Department of Space Medicine, who had identified the altitude limit past which humans could not survive without a pressure suit), the laboratory’s purpose was to validate life-support hardware for a continuously-manned outpost in space and to identify the gravity limit below which human health really began to suffer. Few, however, failed to notice the fact that the station shared a name with that famous hero of the Apollo Program, Neil Armstrong, nor that its mission bore a striking connection to his.

Armstrong began its development as a series of design studies for a permanent habitat module for Spacelab, which would enable the microgravity outpost to transition to a permanent laboratory. While popular with biomedical researchers eager to gauge the effects of very long-term exposure to microgravity, the idea of permanently stationing crew on Spacelab had a mixed reception elsewhere. Some European officials were eager to find opportunities to expand the European astronaut corps, while others (representing materials scientists on both sides of the Atlantic) objected to the inherently greater vibrations induced by a crewed presence on the station, which would compromise the microgravity environment or require extensive redesign of experiment pallets to mitigate the damage. A minority of scientists (primarily those whose experiments had been compromised by mechanical breakdown and could not be repaired for weeks to months until a Shuttle brought a crew up for a visit) were more optimistic about the possibilities for servicing experiments in LEO. The combined effects of all these arguments saw the Space Habitat physically detach from Spacelab, becoming a co-orbiting platform that would maintain position four kilometers aft of the older laboratory.

Once Armstrong slipped the surly bonds of Spacelab’s docking port, it began to attract attention from those elements of NASA concerned with interplanetary missions and the still-novel concept of artificial, centripetal gravity. The first proposal called for a heavy centrifuge to be mounted inside the Space Habitat, to enable testing of mice at lunar and Martian gravity levels. More ambitious proposals followed, calling for an entire habitable section of the spacecraft to be spun up. Though advocates of an all-microgravity setup pointed out that this would eliminate data on long-term human exposure to microgravity entirely, the rotation-advocates pointed out that what data they had already gathered on Spacelab, Shuttle, and earlier programs going back to Skylab and Gemini already indicated that microgravity was a severe hazard to astronaut health, inducing severe muscle and bone atrophy if not counteracted with an intense daily exercise regime (which did not entirely eliminate the problems). Long-term missions to the Moon and Mars would either require artificial gravity in-transit or at the destination to eliminate these problems, or spend the overwhelming majority of their time in a weak gravity field. In light of such considerations, they argued, it was more important to establish the lowest necessary acceleration to prevent the atrophy observed since Gemini.

For a large part of the 1980s, Space Habitat languished in design studies, and the debate between a microgravity habitat and a partial-gravity habitat was hashed and rehashed several times over. It was not until the Wetlab demo flight that the argument began to turn decisively toward the partial-gravity side, as Wetlab demonstrated a method by which partial gravity could be achieved relatively cheaply and easily, by using the discarded S-IVC as a moment arm. Gone were the Von Braun-style wheeled space stations and folding toroids; by 1988, most partial-gravity proposals for Space Habitat called for an advanced derivative of Wetlab, with a docking port coaxial with the spin axis, a dry Equipment and Service Module (ESM) on one side, and the immense Wet Workshop on the other. The design eliminated the difficulty of designing rotating seals between spun-up and despun portions, but set a firm upper limit on the rate of rotation (and therefore maximum acceleration), past which rendezvous and docking for a crewed vehicle would be extremely difficult. Since Wetlab already had a fairly long moment arm, this was deemed an acceptable limit that would still allow tests at Martian gravity levels.

The Space Exploration Initiative suddenly brought the Space Habitat out of the darkness and into the light. As the initiative reoriented NASA from a focus on Low Earth Orbit to long-duration missions to the vicinity of the Moon (and later the Moon itself), the Space Habitat became a crucial tool for developing the long-term life-support technologies and techniques that would be necessary to sustain such missions. By 1990, the consensus at the major NASA centers (Ames, Johnson, and Marshall) had reached an approximation of Space Habitat’s final form--a formation-flying, partial-gravity laboratory co-orbiting with Spacelab. It was this design that entered the NASA budget request for FY1991 as the Partial Gravity Space Habitat, which would finally be named “Armstrong” in the FY1992 request.

Once the spacecraft was defined, the next task was to assign development for the three main components: the Wet Workshop and its associated hardware, the dry-launched Equipment and Service Module (ESM), and the Docking Module. The Wet Workshop contract, naturally, went to McDonnell-Douglas, while the Docking Module was assigned to Grumman Aerospace, which leveraged its experience with the MPEM to build the new pressure vessel. The single largest and most complex part of the station, however, fell to the European Space Agency and its prime contractor, Deutsche Aerospace Aktiengesellschaft (which had inherited ERNO’s work on Spacelab through a series of mergers). Having studied the Spacelab service module (which had been largely developed by Marshall Space Flight Center), DASA set to work combining heritage Spacelab systems with new innovations.

In addition to more advanced computers, the Equipment and Service Module took on a number of improvements to its power generation and life support systems. A number of Russian systems developed for the Mir space station found use aboard Armstrong, including deposition-based CO2 scrubbers (replacing the fragile and constantly-replaced zeolites used on Spacelab). Armstrong’s photovoltaic systems were considerably more efficient than those developed a decade earlier for Spacelab, and were themselves supplemented by a pair of experimental solar-dynamic power systems jointly developed by the Lewis Research Center and the US Department of Energy. Though only providing a fraction of the station’s power requirements (roughly 20% each), the two experimental systems, the Solar Brayton Engine Demonstrator and the Solar Stirling Engine Demonstrator, were a valuable testbed for future power systems designed for the outer solar system, and had a considerably better power-to-weight ratio than the existing photovoltaic arrays. Such systems had been proposed for demonstration on Spacelab, but the need to maintain a vibration-free environment on that station had prevented their installation.

The portion of the Space Exploration Initiative’s hardware which prompted the largest debate was the interorbital tug. The tug was a critical part of Bush's Space Exploration Initiative: the process of returning to the surface of the moon, not just once or twice but to stay, was a vision common within NASA, but it couldn't be accomplished without routine and cost-effective access between LEO and other points in cislunar space. This observation wasn't new: the Space Tug was a concept which had been under study for longer than the Lifter itself, with its earliest ancestors being the tugs which had proliferated in studies during the Apollo era, and formalized in the same Integrated Program Plan which had subsequently birthed the Space Lifter. In these plans, reusable chemical Space Tugs and orbit-only Nuclear Shuttles had served to carry payloads launched by Space Shuttle and Saturn V to the moon, to geostationary orbit, and to staging points for hardware to be assembled for missions to Mars or beyond.

In the mid-eighties, as the Space Transportation System had stepped up to an astounding launch cadence and the Space Shuttle had proved to live up to at least some of the promises surrounding applications for lower-cost manned spaceflight, the Orbital Transfer Vehicle been added back to planning for lunar and Mars missions, now envisioned not only as a cislunar tug but also as a possible booster for Mars missions, using chemical propellants for boosts but in some designs using large heat shields assembled in orbit for more propellant-efficient returns for reuse. By the time the SEI was being laid out, these concepts had been joined by renewed interest in nuclear options, spurred by the military's Project Timberwind development efforts, as well as by new interest in high-power, high-efficiency solar or nuclear electric ion or hall effect systems. Thus, when the Congress approved the funding of the development of an interorbital tug to facilitate the SEI's lunar vision, it was far from clear to any of the congressmen just what they’d bought.

NASA administrators and engineers spent much of 1990 refining the initial concepts of the Lunar Transfer Vehicle even as funding for the program was advancing in Congress. These early studies, funded out of FY1991 funds advanced for spending in 1990, worked to prune down the wide spread of initial concepts into a single vehicle architecture which could be put out for bid. The wildest ideas and those furthest from production were the first to go. Although electric propulsion offered tremendous ability to move large cargoes with minuscule expenditures of propellant, no system of the scale required to fly a multi-ton manned tug had ever been flown. The technology was simply too immature for immediate consideration. The same judgement was also rendered to nuclear thermal rockets. While the Timberwind project had recently found that the technological foundations were strong, the study had explicitly not dealt in depth with the major worries over multi-burn designs, inspection and reuse in space and on the ground, and the surrounding cost structure. In the end, like electric propulsion, nuclear thermal was simply trapped too far down the technology readiness levels to advance onto the critical path.

The elimination of exotic propulsion for the moment meant that traditional chemical solutions would have to suffice, but even within this realm there were a myriad of competing options, from simple conversions of existing stages to modular and drop tank systems assembled in space and refit at the Space Habitat to aerobraking discs and ballutes that would capture back into lower orbits. Here NASA objectives came into sharp opposition, and managers bogged down, caught between lessons learned over the tenure of the development and operation of the Space Transportation System.

The first lesson was the benefit of re-usability and already-built infrastructure to reduce operational costs and protect programs against cancellation. While the expendable programs of NASA's early days, from Atlas to Titan and from Mercury to Apollo had had to fight every year for ongoing approval, the Space Shuttle and Space Lifter had never in almost a decade been in serious danger of cancellation. Other expendable systems illustrated the same: because an expendable vehicle always required new hardware, any interruption in demand could terminate a program or spike costs to the point of cancellation. Even the recent temporary respite for Titan III emphasized this point: the Titans could not be released by the USAF for commercial use since they were pulled from a strictly limited stockpile, and the costs were higher than Lifter even ignoring the storage and stand-up costs. In a break from the Apollo tradition, NASA managers hoped that the lunar Space Tug could be an opportunity to build an infrastructure to reusably and cheaply access the moon. They hoped that LTV and the rest of the SEI could combine to form a sustainable, cancellation-proof infrastructure for access beyond Earth’s orbit, much as the Space Transportation System did for LEO. Such thoughts lead to elaborate schemes, with orbital servicing and long-life aerobraking systems which would not only pave the way for other applications but help justify investment in orbital infrastructure.

Sharing some of the same drive to apply lessons from the STS to the LTV was another faction which drew on the lessons learned more recently during the Magellan accident investigation. This group stressed the importance of long-term inspection and tear down to a reliable reusable vehicle, comparing the reliability of the Space Lifter booster and Space Shuttle glider even after stints of six to a dozen flights between SLIP inspection periods to the expendable S-IVC stage which had failed in the accident. Whenever possible, they stressed the ability to conduct full tear down and refit of reusable vehicles.

While the possibility of extensive orbital hangar infrastructure at the Space Habitat had been an option, the two viewpoints had been in harmony, but the restrictions from a third managerial viewpoint would throw them into a battle only one side could win. This third objection was the cold budgetary realities of the program which Congress had approved: the Space Habitat would not be a massive space operations center where a tug could seek safe harbor for refueling, inspection, and refit, but a smaller orbital outpost aiming to provide mainly biomedical and operational data for long-duration manned missions. Moreover, the part-wet, part-dry single-launch Armstrong Station design that was emerging as the final station architecture was entirely incapable of hosting a hangar, both for size reasons and because of its tumbling end-over-end artificial gravity generation. NASA lacked the funding to even consider a third Space Dock station above and beyond Armstrong and Spacelab that could play harbor to the tug, so any tug would have to return to Earth for any servicing and inspection. Though not every flight, it would still be necessary to disassemble and reassemble any of the more baroque designs to fit them into the Space Shuttle cargo bay at regular intervals, which could easily consume much of the cost and engineering savings over an expendable stage.

Both of the other factions would have to compromise: the aerobrakes preferred by many of the infrastructure designs were all many times larger than the Shuttle cargo bay, and couldn’t be returned. Several studies were conducted on tugs with expendable heat shields, which could be evaluated by the same astronauts which would disassemble and return the more complex engines and tank subassemblies for integration with a new shield and relaunch on Lifter. However, the problems for the overall architecture were serious, and the designs were largely rejected from consideration, though large one-use shields remained in consideration for other applications like Mars. For their part, the reliability faction had to give up extensive hands-on inspection by astronauts every flight. In exchange, they would get regular inspections on the ground, in ordinary clean room conditions instead of whizzing along at kilometers per second in a space hangar. What emerged was a new architecture: a chemical modular architecture, with all inspections and overhaul performed on the ground and minimal manned turnaround support in space. Though modular designs were acceptable to meet this goal, their assembly had to be minimal, and they should be reusable to the maximum extent practicable. The result was that the LTV design process shifted from a wide-open design space to a more restricted, perhaps nearly non-existent one. The design had to have subassemblies small enough to fit inside Shuttle for return, while also being simple enough to cost-effectively tear down for return. Under such limits, the replies to NASA’s request for proposals when the LTV was put out for bid had to be unorthodox to meet the constrained trade space.

The design for the LTV was finally put out to bid in 1991, and the resulting proposals were incredibly diverse. Several companies submitted multiple proposals, with each testing the bounds of one or another of the design ground rules. The highest-rated Lunar Transfer Vehicle designs, upon analysis, were in many case those that strayed furthest from assumptions about what the Space Tug would have been only a few short years before. A number discarded the traditional assumptions of hydrogen/LOX propellants and instead considered a variety of hydrocarbon propellants, mixing LOX with methane, propane, and kerosene, or even using alternate propellant pairings like Grumman’s studies of kerosene/hydrogen peroxide. Initially selected for higher storage densities for smaller tanks, many combinations also yielded pairings more easily stored on orbit than the traditional but evanescent hydrogen. The designs of modular tanks led several firms to consider not just a modular tug, but a multi-stage one--the traditional solution to getting more performance out of limited mass.

The winning proposal, by McDonnell-Douglas, added a novel approach to aerobraking to these basics. McDonnell-Douglas engineers realized that traditional aerobraking heatshields had to be so large due to the assumption of a one-and-done braking maneuver--a dive deep enough and hot enough to scrub off all the vehicle's excess velocity in a single pass, just as a shield designed for an orbital entry did. To dissipate the large heat pulse of such an entry, the shield had to spread it over a large area and use exotic materials. However, if the braking was spread over multiple perigee passes, it would have correspondingly shallower dives into the atmosphere and lower heat input. Studies McDonnell engineers discovered in archives from the 1970s had found this sufficient passes could yield heat shields which didn't require extensions, but could instead be wrapped around the body of a traditional stage which fit within the Shuttle cargo bay. A small "flare" to stabilize the stage would have to be disposed of on each service mission, but the entire heat shield of the "small brake" design could be built of less exotic materials and still offer aerobraking’s benefits to weight.

McDonnell’s final proposal was wildly divergent from the original 1990s concepts, but it almost wrapped back around to conventional: a cylindrical stage with a docking collar on each end just barely capable of fitting within the Shuttle cargo bay. Two stacked atop one another and fired in series were capable of sending a substantial payload into lunar orbit. Interest from Mars-focused McDonnell engineers and density impulse trades lead to their proposal calling for a new, methane expander-cycle engine, sized such that a cluster of four could fit around the aft docking collar. Even so, the engines' fit was tight, requiring them to gimbal out to allow the docking port to extend for assembly and gimbal in to clear the shuttle bay doors on return. Pratt & Whitney, the manufacturers of the RL-10 expander engine since its introduction, were consulted heavily on the design for this new expander cousin. A heat-shield would close over the payload end of the stage for aerobraking passes, with the engines protected by the stabilizing flare, which would be discarded before return to Earth for inspections when necessary. To further reduce heat shield loading, the hottest portions of the nose would be actively cooled with methane during entry. The four-day return to LEO from lunar transfer orbit for the second tug and two-day return to LEO for the first tug would slightly reduce the maximum tug flight rate, but also meant that the actively-cooled heat shield should be indefinitely reusable, though the niobium alloy selected could manage multiple complete missions with purely passive heating as an emergency measure.

The tugs were interchangeable: any tug could be a first stage or a second stage for a lunar mission or be used independently with a partial propellant load for geostationary orbit missions. Two or more tugs could be launched together on a single Lifter with a partial propellant load, and attachments allowed for propellant transfer to top them off in orbit, even using other tugs as tankers. Their methane/oxygen propellant and low-pressure expander engines meant there was little to go wrong: they could last fueled in orbit nearly indefinitely, and the engines should have lifespans similar to RL-10s, which had lasted the equivalent of hundreds of tug missions on the test stand. Even so, the entirety of any damaged or worn tug, less the flare, could be returned to Earth with a single Shuttle mission, which could also carry up a replacement fresh from refit. NASA selected McDonnell’s proposal in August 1991. A follow-up contract went to Pratt & Whitney for the design and test of the required Lunar Transport Main Engine, proving their early design contributions to be worthwhile. Although small enough that Pratt & Whitney’s early full-scale mockups could fit in a car seat, the expectations on the engine were high: it would have to live up to the reputation of its cousin, the expander-cycle RL-10, and bear the schedule pressure of the entire lunar program on its tiny thrust mounts. It was, in point of fact, the first major clean-sheet engine development project in the United States since the cancellation of the SSME project two decades prior. Pratt’s victory was contested by Aerojet and Rocketdyne, who argued their relationship with McDonnell during the design process made the contract award unbalanced. Aerojet was eventually satisfied by winning the contract for the LTV’s methane/oxygen thruster system, while Rocketdyne would have to content itself with its unquestioned dominance of every other STS engine. Only TRW, whose pintle engine design they hoped would have more applicability in future alternative roles, was entirely shut out. Still, by the end of 1991, the design of the tug had been settled and planning was in work to hopefully see it fly by the notional 1996 date. NASA planners could finally turn their attention to how it might be used to implement the longer term goals for lunar orbital stations and for a renewed series of lunar landings, while engineers at McDonnell and P&W set to work beginning to turn the designs of the LTV and the LTME into reality.

The decision to use chemical propulsion for the LTV did not mean the end of electric thruster development. Since the 1960s, electric propulsion had been a favored topic of research on both sides of the iron curtain. In the Soviet Union, these efforts bore fruit in the form of Hall-Effect Thrusters that, since 1971, have provided Soviet and Russian satellites with reaction-control capability. Using far less propellant than chemical thrusters, these HETs greatly extended Soviet satellites’ operational lifespans. The long lifetimes of HET-equipped satellites made the technology particularly attractive for American and European commercial satellite manufacturers, and made HETs one of the first advanced Soviet aerospace technologies to cross the ruins of the Iron Curtain. Following the launch of a small demonstration satellite (sponsored by MIT and the Space Studies Institute), GeoStar began designing HETs into its next-generation communications satellite design. Ford Aerospace, not to be outdone, also equipped its FS-1300 geostationary satellite bus with Hall-Effect stabilization technology, though they did not deploy the first so-equipped satellite until 1998.

While the Soviet Union focused on Hall-Effect reaction control for earth-orbital applications, American researchers set their sights further outward. From the 1960s, researchers at Lewis Research Center had imagined that their low-thrust propulsion systems would power eventual human missions to Mars and unmanned probes to the outer planets. Perhaps for this reason, the American low-thrust propulsion effort failed to attract as much official support as its Soviet counterpart; after two demonstration missions in 1964 and 1970, the American effort mostly went dormant until the late 1980s. When it was finally revived, the effort began, not at Lewis Research Center, but at its old rival the Jet Propulsion Laboratory. It was there that a team of engineers proposed, in 1987, to launch a pair of low-mass, low-thrust, low-cost spacecraft to the Moon. Dubbing their proposal “Lucky 7,” after one of them turned a presentation slide with “L L L” written on it upside-down, the team designed a 150-kg spacecraft bus equipped with two xenon-propelled Gridded Ion Thrusters. Designed to fly all the way from Low Earth Orbit to a polar Low Lunar Orbit, each Lucky 7 spacecraft would carry one instrument. The first, a gamma-ray spectrometer (left over from the instruments built for the Apollo 19 mission), would be designed to detect the presence of heavier elements (phosphorus, potassium, and others) in the lunar crust, while the second, a neutron spectrometer, was designed to characterize the presence of volatile elements in the Moon’s crust. Together, the two spacecraft would deploy from a Space Shuttle payload bay and slowly thrust toward the Moon, reaching their target orbit two years after launch. In the process, they would demonstrate the operation of a solar-electric thruster in near-earth and cislunar space, proving a concept for solar-electric tugs from Low Earth Orbit to higher destinations and measuring the exact impact of the Van Allan radiation belts on spacecraft slowly crawling through the dense, energetic belts. The actual scientific observations they’d perform at the Moon, compared to this value, were an afterthought in the original JPL presentation.

Lucky 7 floundered for a time at NASA. The Jet Propulsion Lab was more concerned, in 1987, with planetary science, particularly that performed by very large, very high-budget missions like Galileo and the Mariner Mark II series. Lewis Research Center, for its part, was open to the project, and the JPL engineers found much support among the Space Shuttle Program Office at JSC, who were always happy to find a payload to fly in the Shuttle payload bay. NASA headquarters, however, was reluctant to allocate funding for the small program, struggling as it was to get the Mariner Mark II program funded. It was not until the announcement of the Space Exploration Initiative that Lucky 7 (by 1990, given the more dignified name of “Inter-Orbital Navigator,” or ION) suddenly found a wave of new support from NASA’s leadership. Influential voices at NASA headquarters, with an eye on long-term missions to Mars, had been proposing a solar-electric tug (with a heavily-shielded payload bay) for years to transport payloads to geostationary orbit or to earth-escape. Between SEI and the panels before it that called for NASA to develop such a reusable in-space propulsion capability, government support for such a vehicle had never been higher. ION went, at least briefly, from a cheap technology demonstrator to a critical step on the path to the Moon and Mars.

This privileged status was not without its drawbacks. When the astronaut office pointed out that spending months to years in the Van Allen belts or even the lesser interplanetary radiation environment would have a deleterious effect on crew health and morale, making a high-thrust chemical tug necessary anyway, NASA’s reference Moon and Mars missions abandoned the solar-electric tug, instead distributing more flights onto the chemical tug. ION went from a critical step on the path to the Moon to a critical step on an unknown path. Though the MIT and SSI work in bringing Russian Hall-Effect Thrusters over to the West had helped bolster some support for low-thrust propulsion research at Lewis (which designed a cislunar tug with chemical main propulsion but electric station-keeping capability), the prevailing momentum toward more use of commercial hardware at NASA had ironically come to favor hardware developed in the Soviet Union.

Equally ironically, ION’s saving grace turned out to be its meager scientific payload. Just as the Apollo program was preceded by Ranger and Lunar Orbiter and Surveyor, so the new return-to-the-Moon would be preceded by a small fleet of unmanned spacecraft to characterize regions of interest in those parts of the Moon far from the Apollo landing sites. Though JPL had already drawn up plans for a Lunar Observer spacecraft, to go with the Mars Observer planned for launch in 1992, ION, as a much smaller set of spacecraft, could be ready to fly by the end of 1991, thus reaching lunar orbit by 1993, several months before Lunar Observer. The two spacecraft would provide additional data on the contents of the lunar crust, and simultaneous tracking from Earth would allow scientists to collect high-quality data on mass concentrations in the crust, allowing them to map the notoriously “lumpy” lunar gravitational field, aiding planners for future lunar orbital missions and aiding geological studies of volcanic activity on the Moon. This argument won the day at NASA, and a pair of spacecraft originally designed to test technology, whose cosmic sensing instruments were almost cosmetic, were saved through the intercession of geologists.

The 1990s also saw advances in propulsion below the Karman line. The European effort to replace Ariane took a sharp turn for the exotic with ongoing work in Britain on their Horizontal Take-Off and Landing (HOTOL) project. Based on the work of British engineer Alan Bond, the HOTOL project proposed to use pre-cooled jet propulsion combined with a rocket engine cycle to launch an aircraft directly from the runway to orbit. However, looking at the experience of the American Aerospaceplane project of the 1960s, British Aerospace was hesitant to commit to a full development effort for a single-stage-to-orbit vehicle, preferring instead to leverage the engine technology for a first-stage, horizontal-landing booster. British Aerospace entered its own air-breathing design into the argument over how to loft a reusable core stage off the ground. The HOTOL design submitted called for a runway-launched air-breathing booster that would propel an upper stage to Mach 7, while liquefying enough atmospheric oxygen to fill up an upper stage. The upper stage would be propelled by a fairly conventional rocket--the most revolutionary aspect would be that it separated from the booster heavier than it launched off the ground.

The HOTOL proposal was both reasonable, economical, and exciting. It promised full-reusability on a two-stage launch vehicle, with a technology that was just close enough to be feasible. Compared to the cautious conservatism of the Ariane design, it was a breath of fresh air, one that was appealing to engineers across Europe. In 1984, following more detailed design studies by Rolls Royce on the liquid-air-cycle engine proposed for HOTOL, Arianespace officially partnered with British Aerospace to develop the HOTOL launch vehicle, with British firms developing the reusable booster and continental European firms developing the upper stage and hydrogen-burning main engine.

The new booster design was comparatively smaller than the Space Lifter architecture, aiming for a maximum of 16 tonnes to Low Earth Orbit. Among other payloads, this would enable the new RLV to lift an expanded manned capsule based on the Hermes capsules in development for the lunar program, and the small size would enable reduced mission costs. In observing their American counterparts, Arianespace managers noticed that satellite operators were often reluctant to share their rides to orbit with other companies, as the logistics of arranging the shared launch tended to induce delays as suitable partners were found. As a result, they aimed to keep launch costs low enough that single-satellite launches would be economical. Though the Ariane replacement system would not be ready until around 2000, its potential to reduce the cost of space access added to the wave of investment that TPLI and the planned Low Earth Orbit constellations drove in the early 1990s, spurring ever more grandiose plans to use the reusable systems to their full capacity.

The designers of technology demonstrators like Inter-Orbital Navigator probes and the new European efforts in airbreathing space launch systems hoped their technologies could one day contribute to the broader exploitation of spaceflight. For the moment, however, they were just following in the shadow cast by the Space Lifter and the new projects started to fuel the Space Exploration Initiative’s return to the moon. They held promise, but the center of gravity of advances in spaceflight lay beyond their development horizons. They were scarcely less remote from the action than the general public. Though the data these developments returned would be of value later, their immediate interactions with the return to the moon was limited to justifying budgets and grants while tracking the progress in trade papers of the vehicles which would, should they succeed, carry humans farther than they had gone in decades.
 
Last edited:
sloshing in gutters on the former sides, now the bottoms of her vast propellant tanks.
Ah, so they decided to use smaller dedicated tanks.
Space Habitat physically detach from Spacelab, becoming a co-orbiting platform that would maintain position four kilometers aft of the older laboratory.
Huh. Interesting approach. Does vibration reduction really require a separate station?
The same judgement was also rendered to nuclear thermal rockets. While the Timberwind project had recently found that the technological foundations were strong, the study had explicitly not dealt in depth with the major worries over multi-burn designs, inspection and reuse in space and on the ground, and the surrounding cost structure. In the end, like electric propulsion, nuclear thermal was simply trapped too far down the technology readiness levels to advance onto the critical path.
Really? It's been in on and off development since 1959, and gone through multiple test firings: I'd expect it to be a potential future upgrade, at least.
new, methane expander-cycle engine, sized such that a cluster of four could fit around the aft docking collar
Well that's new. Is this the first entirely non-OTL engine design you've put in?
It was, in point of fact, the first major clean-sheet engine development project in the United States since the cancellation of the SSME project two decades prior.
Well, if it's clean sheet TTL, and I haven't heard of it OTL, I guess it is original. Interesting concept, I had never heard of a methalox upper stage until ITS, and never on its own.
This argument won the day at NASA, and a pair of spacecraft originally designed to test technology, whose cosmic sensing instruments were almost cosmetic, were saved through the intercession of geologists.
Lovely little bit with ION. Captures the tangled development paths of NASA projects very well.
The European effort to replace Ariane took a sharp turn for the exotic with ongoing work in Britain on their Horizontal Take-Off and Landing (HOTOL) project. Based on the work of British engineer Alan Bond, the HOTOL project proposed to use pre-cooled jet propulsion combined with a rocket engine cycle to launch an aircraft directly from the runway to orbit.
Well then. Brits are going full throttle.
The HOTOL design submitted called for a runway-launched air-breathing booster that would propel an upper stage to Mach 7, while liquefying enough atmospheric oxygen to fill up an upper stage. The upper stage would be propelled by a fairly conventional rocket--the most revolutionary aspect would be that it separated from the booster heavier than it launched off the ground.
Very ambitious. I fear this will become TTL's Shuttle.
 
Top