Post 0: Introduction Image
The Dream Survives
The Dream Survives
- Thread starter e of pi
- Start date
Post 1: Climbing the Wall
Climbing The Wall
The Shuttle is to space flight what Lindbergh was to commercial aviation.
— Arthur C. Clarke[1]
The Shuttle is to space flight what Lindbergh was to commercial aviation.
— Arthur C. Clarke[1]
If the Space Shuttle was to live up to that promise, it would have to demonstrate and then safely sustain the flight rate demands of becoming the primary American launch vehicle. This would require flying a wide range of scientific, military, and commercial payloads from both national and international customers. Even with some smaller payloads combined, this demand could reach dozens of flights a year. The Shuttle program would have to reach this rate not just once, but to sustain it for year after year. Since its debut, the launch rate had increased steadily from two in 1981, three in 1982, four in 1983, and five in 1984. The nine flights in 1985 were a marked improvement not only in the total but in the rate of change. While the average of near-monthly launches in the latter half of 1985 was a major achievement, it would have to be exceeded radically in 1986. With four orbiters online, NASA contemplated a crowded flight schedule for the year to come, with fifteen launches planned that year. On the schedule were numerous time-sensitive missions, heavy-payload missions, and the debut from Vandenberg which would push Shuttle to its limits. For the start of the year, though, there were merely two launches made special only by their passengers: a comsat mission STS-61-C carrying Representative Bill Nelson, and a Tracking and Data Relay System (TDRS) satellite launch on STS-51-L carrying teacher Christina McAuliffe.
The launch of Challenger on STS-51-L was regarded as largely routine. The crew for the flight had been announced in January almost a year prior, but Christina McAuliffe wasn’t added to the crew until July with only six months to prepare to teach students from space. On January 15th, 1986, the Launch Readiness Review was conducted and signed off on the mission as ready to fly. A week later, on January 22nd, the mission was delayed from the originally planned launch of January 23rd to January 26th. The day before the revised date, however, weather issues in the forecast delayed the launch to the 27th. The weather continued to plague the mission, though, as issues with abort landing winds and an exterior hatch handle scrubbed the attempt on the 27th. That night, Thiokol engineers reviewed telefaxed data from Kennedy and requested an evening meeting to discuss concerns about solid rocket booster seals in the colder temperatures predicted overnight and the next morning. Meetings went back and forth for hours between 5 PM and 8 PM. Thiokol’s engineers recommended against any launch in temperatures below 53 degrees Fahrenheit. However, with the mission having been delayed several times already, ultimately the engineers were overruled by both NASA and their own managers. Thiokol was pressured into giving a recommendation of “acceptable to proceed” for the next morning. Knowing nothing of these concerns, the STS-51-L crew boarded Challenger the next morning and were seated aboard OV-099 by 8:36 AM. At 11:15 AM, the inspections for ice at LC-39B concluded, a pad which had not seen a flight since Apollo-Soyuz. The launch team gave the GO, and the count continued. At thirty-seven minutes and fifty-three seconds past eleven in the morning, Challenger’s main engines ignited. Seven seconds later, the solid rocket boosters ignited and the vehicle wrenched itself from the pad.
Barely a second into the launch, a puff of smoke was seen emanating from the Center Field Joint on the right SRB, the benign specter of a much more vicious problem: the total blow-through of the field joint, with hot exhaust gasses flowing through both O-rings on the SRB like a blowtorch. However, a second later, the plume stopped as a chunk of burning fuel grain slapped across the hole in a tenuous plug. The grip of that plug would decide the fate of the mission. In the third second of flight, the Public Affairs Officer enthusiastically proclaimed to the eyes of the children and other audiences around the world, “Liftoff of the 25th space shuttle mission, and it has cleared the tower!”
Seven seconds into the flight, the orbiter’s world flipped as the stack began the roll program to rotate “head-down”. The vibrations and rocking of the engines fought against the chunk of debris, stuck in the gap like the story of a boy’s finger in the dike. Despite the shaking, the combustion pressure of the motor itself kept it locked against the motor’s field joint. As the orbiter aligned itself, the commander and pilot could feel the weather in her movement. Pilot Michael J. Smith noted for CAPCOM, “Looks like we've got a lotta wind here today.” At 35 seconds into the flight, the orbiter’s main engines throttled back to 65% for the “thrust bucket,” a programmed reaction to reduce the aerodynamic pressure on the vehicle as it approached the peak known as “max-Q”. As Challenger cut into this peak, it found for a moment a patch of clear air, like the eye of a hurricane in the gusts buffeting the vehicle in its ascent from the ground. It was like everything was still except the roar of the solid boosters, the howl of the SSME, and the rattle of structures and internal fittings in the crew cabin. The three Space Shuttle Main Engines throttled back up, and the ground breathed a sigh of relief at the call from the commander: “Challenger, go at throttle up.”
As the power came back up on the three main engines, the pressure in the solid rocket boosters reached its peak, and then began to fall off. In the right-hand booster, the solid fuel plug hung in place, vibrated, and then stuck fast to the gap again. It was a knife’s edge of safety, enabled by unusually smaller attitude response rates. Finally, at 11:40 AM Eastern time, the pressure in first the right and then the left booster fell below 50 PSIA, triggering SRB separation. The expended boosters fell away from the Shuttle, arcing up to an apogee of 220,000 feet. Nearly four minutes later, the drogue parachutes deployed to stabilize their fall through the atmosphere, followed by the main parachutes at 11:44:28. Finally, at 11:44:41 Eastern, the right-hand SRB impacted the water at the designed rate of 76 feet per second under full parachute canopies. The equivalent of a 53 mile per hour car crash, this was normal for the boosters. Less than two minutes later, Challenger’s engines cut off to place STS-51-L into orbit. [2]
The tremendous public success of the STS-51-L mission in outreach to the average American was difficult to measure — for many who had been children, the live stream of the launch would be a defining memory. However, in reviewing the flight data, NASA quickly discovered not all had been well. The slug of partially burnt fuel material which had apparently plugged the breach was dislodged at some point after separation, though reports couldn’t determine if it was the drogue deployment, the main parachute deployment, or the final 53 mile per hour impact with the water. When hauled out of the water and broken down, charring on the booster field joint showed clear signs of a blowtorch-like total burn though. With this clue, NASA was able to find the brief plume of burning gas in the video of the launch. If the plume had not been plugged incidentally, and had instead been turned towards the vehicle and critical hardware…. [3]
The risk of the vehicle’s loss drew black stares into the middle distance around engineering data review conference tables. NASA suddenly found they had much more time to review Thiokol’s suggestions around temperature limits for Shuttle launches. The capture feature design considered for the filament wound boosters was now to be included in all future standard casing orders as well. Older casings were to be phased out of use as news sets became available, and were only to be used with caution in any weather extremes.
Inside of NASA, STS-51-L raised eyebrows at the risks of so-called “ordinary” missions. However, there were riskier ones on the horizon which drew notice away from an already-complete routine TDRS deployment for those outside the SRB team. The next three launches of the year would all be dictated by the schedule of the spheres, not NASA. Columbia was signed up in March for a mission to observe the peak of Halley’s Comet, STS-61-E. Then, in May, loomed the pair of missions known to their crews as the “Death Star” flights--the launch of the interplanetary probes Ulysses and Galileo with the Centaur-G Prime rocket stage riding in the Shuttle payload bay. The launch of STS-61-E was complicated mostly by its extraordinarily tight turnaround, with the nominal liftoff date of March 6th just 47 days after the orbiter Columbia’s previous landing at Edwards Air Force base, and 53 days after her previous launch. According to mission planners, the peak of the possible scientific return was on March 10th. If the launch slipped to March 20th, it was almost not worth flying. Technical staff at Kennedy Space Center were strained by spare parts shortages and on-the-job training of new personnel to make up enough hands for the four active orbiters. While Shuttle’s ground processing team worked miracles, the flight crew struggled to find time in the pair of temperamental simulators available in Houston. In the end, the task was managed despite a two-day weather delay. The vehicle’s liftoff on March 8th 1986 missed the previous record for fastest launch-to-launch turnaround by under 8 hours, and the flight crew spent nine frantic days in space collecting every morsel of data they could manage trading Red Shift and Blue Shift operations around the clock.[4]
The harried crews had April off from launches, but expecting a break was a cruel prank. Atlantis, laden with her Centaur and the Galileo probe, was on the launch pad at 39A from the start of the month, and only a few weeks later Challenger would take up her position on LC-39B with her own Centaur and the Ulysses probe. STS-61-E had been a race, but this would have to be a ballet of synchronicity as pad crews worked to have both missions ready to go in time for the trans-Jupiter departure window both probes needed. For the first time, live liquid stages were integrated with the Shuttle orbiter. Astronauts and ground handlers sweated over the thin skins of the Centaur balloon tanks in their cradles. The propellant load paled next to the external tank a few feet away, but if anything went wrong this tank was contained right in the core of the Orbiter’s critical systems. Worse, the Centaur G-prime stage and the probes it would launch had been designed for payload figures promised early in the Space Shuttle program. For the heavier mission, Galileo, the Centaur and the probes totalled 25.2 metric tons. [5] In fact, as-built, the orbiter could only reach that figure with the combination of a stripped-down cabin configuration and minimal crew, an unusually low parking orbit altitude, and with its three main engines pushed from the standard 104% to 109% throttle.This was a change which sounded minor, but came with radical increases in stress on the turbomachinery and cooling systems for the engines. With so much that could go wrong and the requirement to manage a pair of flights within just four days, the two launches gained the nickname “the Death Star launches.”
The pressures of getting two launches scheduled off the pad within four days was always unlikely, and the flight planning reflected this. The lighter of the two probes, Ulysses, was planned to launch with the heavier orbiter, Challenger, but would still have a slightly longer window. Thus, it was scheduled first in the 21-day opportunity, leaving the lower energy middle of the window for the more challenging launch of Atlantis and Galileo happening second. The astronauts and ground crews, used to the standard procedures for the Space Shuttle, were troubled by some of the approaches brought by the Centaur team to working with their thin-skinned vehicle, no matter how successful those approaches were on uncrewed flights. For the flight crews, the missions were almost boring other than the launch risks. Challenger’s hard-pressed SSMEs muscled the orbiter off the ground one day behind schedule, and the crew breathed a sigh of relief as they entered orbit. It became a race against time they could control. The Centaur would have to separate and conduct its burn within nine hours of launch to manage boiloff, with the first window seven hours into the mission. The new Centaur stage proved well-behaved as any Payload Assist Module in its deployment. Challenger cut Ulysses and its Centaur loose on schedule, monitoring the burn. The crew celebrated the Centaur stage’s burnout with a reading to the ground of a passage from Tennyson’s Ulysses:
With Ulysses having pushed off and the crew aboard Challenger sitting well in order, all that remained was waiting the days needed for an on-time return to the ground. In the meantime, the crew conducted a cut-down list of orbital experiments on the mid-deck, the better to get the orbiter back on the ground and clear the skies for Galileo’s own explorations of Jupiter. The first of the “Death Stars” had flown…but all too many Star Wars fans could attest from 1983’s Return of the Jedi that Death Stars came in pairs.
Clouds were gathering for Atlantis’ launch even as Challenger orbited, both literal and metaphorical. The weather at the Cape had closed in, preventing the 1-day turnaround from Challenger’s landing to Atlantis’ launch. Moreover, the world had awoken to new worries about nuclear power like Galileo’s radioisotope thermal generator (RTG). Just weeks earlier, the news had come from the Soviet Union that, “There has been an accident at the Chernobyl Nuclear Power Plant. One of the nuclear reactors was damaged. The effects of the accident are being remedied. Assistance has been provided for any affected people. An investigative commission has been set up." Challenger’s launch just ten days after the disaster’s announcement came so close that little new organization could be managed by anti-nuclear protesters, when it wasn’t even clear in the west the true scope of the accident. Still, coverage of one probe’s launch stressed the risks and the impending launch of another. Protestors swarmed Galileo launch, even as behind the iron curtain men and machines worked to quench the fires in the bleeding heart of a wounded nuclear reactor.
Engineers and technicians attending to Atlantis on the pad every morning drove past a line of signs and banners until KSC’s main gates, only to confront the regular problems of the Shuttle program: too much to do, too little time, too many tasks, not enough spares, and the capricious weather of Cape Canaveral’s summer heat and rains. Mikhail Gorbachev took to television on May 14th to assure the world that “the worst is behind us” but the Space Shuttle program was still in the trenches. The launch, originally scheduled for May 15th, finally slid to May 18th due to weather, only to be delayed one more day by a scrub over concerns about excess hydrogen buildup on the pad. On the morning of May 19th, 1986, Atlantis’s three main engines lit, followed by her SRBs, and for the fifth time that year the skies of Florida were alive with the crackle of Shuttle engines.
Five seconds after launch, plans began to fall apart, not due to Centaur, or nuclear disasters, but due to simple electricity and computers. A short on one AC bus took out one primary and one secondary controller on each of two main engines. The loss restricted telemetry available to the ground, limiting the insight of flight controllers into the operating regime of the engines. This failure would never have mattered were it not for two other critical events. The first, and more benign, was caused by a sensor issue on the left engine. With only one engine controller, a biased sensor in the remaining controller led to a slight tendency to run fuel rich, burning more hydrogen than optimal. Such issues with mixture ratio weren’t unheard of. Every flight carried reserves of both hydrogen and oxygen to ensure that the orbiters could meet their mission needs while protecting against physically running out of propellant because of slosh, sensor irregularities, or engine underperformance. Indeed, the risk of a “dry” turbopump overspeed was high enough that quantity warnings would automatically shut down engines which were not seeing sufficient flow into their manifolds, with underperformance being better than the risk of uncontained turbomachinery failure.
The second and much more important event was a single gold pin used to plug an injector that was ejected out of the right engine as it started up. This pin shot like a round from a rifle, and was nearly cleanly ejected out of the engine bell when it nicked three of the tubes that made up the nozzle extension. These tubes, which circulated hydrogen before it was burned in the engine, were the only way that the engine could repeatedly withstand the high temperatures of combustion. If as few as five adjacent tubes failed, the nozzle would eventually overheat and break. This failure would have promptly resulted in the loss of the spacecraft, and her crew. Because the nicked tubes were downstream of the flow sensors on the engine, this failure could only be inferred from the reduction in hydrogen that was being delivered to the engine. The engine controller’s simplistic logic, though, could only assume all hydrogen delivered to the engine was being burnt there, not that it was leaking overboard. In response to the extra hydrogen needed, the controller also increased oxygen flow to keep up performance.
Both hydrogen and oxygen were now leaking overboard in large quantities, though ways largely invisible to the ground controller’s limited telemetry except in increased turbine temperatures. The failure of the right engine’s sensors causing excessive hydrogen use was in many ways a boon--with the left engine demanding more oxygen, every second it was running brought the shuttle closer to running out of oxygen prematurely. If all the oxygen was consumed but thousands of pounds of hydrogen stayed aboard, the excess weight could leave the highly-demanding mission doomed to an underperformance. An underperformance could be anything from an abort-to-orbit which might imperil the ability to reach Jupiter with the probe to the riskier trans-atlantic abort with a Centaur and nuclear space probe in the payload bay, putting Johnson, JPL, and Lewis’ performance in designing for contingency to the test.
On the ground, the flight director and their controllers worked the problem through the remaining rudimentary telemetry. With the engines straining at 109%, their performance had been of concern in planning, and the Booster backroom’s main engine controller was watching for any of a laundry list of cues for serious issues. Temperatures on the turbopumps on the left engine were elevated, but even at 109%, not enough to demand a shutdown. The right engine’s issues almost escaped concern, other than trying to work the failed telemetry, while the focus remained on the left through the “throttle bucket” of Max-Q and the call to throttle-up. When the engines responded and trajectory officers confirmed no underspeed on the vehicle performance, Booster gave the Flight Director the go-ahead to press on. As the shuttle climbed and the two competing issues struggled to burn through the orbiter’s margins on the propellant load, the controllers checked off the potential failure modes: clear for single-engine pressing on to trans-Atlantic abort. One engine press to orbit. One engine press to main engine cutoff (MECO).
Finally, there were no contingencies left to take: as the quantities bled through the engine’s prevalves, there was nothing to do but to press on to MECO and take whatever orbit resulted. Seconds before the trajectory commanded shutoff, though, the dreaded warning flashed through the system: liquid oxygen level low! The prevalves shut to protect the turbopumps, and the engines shut off, even as the telemetry flashed down to the ground. While the booster team worked through engine shutdown and preparation for ET separation, the flight dynamics controllers evaluated the results of the burn. Weighing the hydrogen leak against the oxygen leak, the oxygen shutoff had left the vehicle a bare 13 meters per second underspeed — close enough to nominal to proceed with the mission, and indeed low enough that no immediate orbital maneuvering system (OMS) burn was needed to correct it.
As trajectory planners updated the Centaur deployment strategy and briefed Atlantis on the results of the ascent, the propulsion team summed up the hectic end of the burn with the Flight Director:
Flight: “Booster, did you see the flash on the low level sensors prior to MECO?”
Booster: “Yeah, we saw it--all four low just before shutoff. With the underspeed, it looks like a LOX low level cutoff.”
Flight: “Yikes!”
Main Engines: “You bet.”
Booster: “Concur.”
Flight: “We don’t need any more of these.“[6]
The flight director offered a decent summary of the mission’s issues and the program’s struggles for the year to date. The Shuttle had carried its heaviest payload to orbit yet through challenges worthy of a simulation supervisor’s most devious plans, but yet again a Space Shuttle mission had scraped the ragged edge of danger. Atlantis deployed Galileo and the second Centaur-G Prime stage, with the maligned Centaur proving the most reliable element of the mission. Its performance was right down the middle of the tolerance bands as it sent Galileo on its way. Now, there was a month before the next flight and the only three time-critical missions of the year were already completed. The program finally had a moment to catch its breath and evaluate problems from the first half of the year and make plans for the second
Ongoing shortages in spares and technicians had haunted the program, leading to routine swapping of components among the fleet: body flaps, elevons, even entire landing gear assemblies were swapped to whichever orbiters were “head of the line” to give the cast-off assemblies time to be repaired properly. More orbiters flying meant more time between missions if the flight rate was constant and more parts to raid, but insufficiently trained technicians left pad rats and orbiter processing facility personnel scrambling to keep up with the increase in flights. Something would have to give, and give it did. Not long after the twin Centaur flights, NASA announced that due to handling delays and reviews with Thiokol about the new Filament-Wound Composite (FWC) solid rocket boosters, Discovery’s debut from Vandenberg Air Force Base’s SLC-6 would be delayed from June to later in the year. The time allowed Discovery to serve as her own Flight Readiness Firing test vehicle at the new pad, saving the need to fly another orbiter out after Enterprise’s service for fit checks and integration tests. The FWC boosters were given a full go-over by Thiokol and NASA, in part to evaluate their use at the Cape. The new booster’s 5,500 lbs increase in Cape payload would help mitigate the need for the 109% throttle setting on Centaur missions in the future, answering another of the program’s 1986 concerns.
Moreover, the Hubble Space Telescope was experiencing its own struggles with flight software. The launch had already slipped from August to October by February. In June the Hubble team determined they wouldn’t be ready to ship the telescope to the Cape in 1986 at all. No additional payloads were available to move up in the flight order at such a late date, unlike the Earth Observation Mission 1/2 which had taken the planned August slot on Atlantis from Hubble in February. This opened up precious margin for a Shuttle launch tempo already accumulating slips. Columbia and the STS-61H launch of the first British and Indonesian astronauts, overseeing communication satellite deployments for their countries as well as a routine launch of another satellite for Western Union, was seven days late with its launch on July 1st. Challenger’s TDRS launch for STS-61M on August 5th was two full weeks late, and when Atlantis finally hauled Spacelab and the EOM-1/2 equipment to orbit on September 6th, it was in turn nineteen days late. The impacts snowballed, and with limits on parts and personnel, there was little ability to make up delays. Twice in the previous year the program had managed individual orbiters making launch-to-launch turnarounds of under 55 days. In the second half of the year they were considering an orbiter lucky which made it to flight within 100 days of the previous flight. Turnarounds over 110 days outnumbered those under 75.
The year did close out with a number of notable successes: flying two different Department of Defense payloads within two weeks of each other, one aboard Columbia and STS-61N from the Cape on September 25th, the other with Discovery’s debut from Vandeberg on October 5th. Aboard the latter was not only an Air Force “Manned Spaceflight Engineer” as payload specialist, but also Edward “Pete” Aldridge, the Under-Secretary of the Air Force and head of the National Reconnaissance Office. Like Senator Garn and Representative Bill Nelson before him, Aldridge got to put his money where his mouth was, and his butt where his money was. In orbit, the mission’s commander pinned on the gold wings of an astronaut.
The debut of a new pad and new solid rocket boosters became another political junket to curry favor with the NRO. These alliances with NRO and other agencies were part of NASA’s resistance to attempts by the Department of Commerce to restrict Shuttle commercial payloads and charge higher prices. In an effort to encourage commercial use of other legacy launchers like Atlas, Titan, and Delta, the Department of Commerce hoped to force NASA onto a “level playing field” by including some or all the of the base cost of the Shuttle program with the prices charged to each additional Shuttle launch and to restrict the number of payloads available for Shuttle. NASA’s point was that the government would have to pay Shuttle’s base costs regardless of whether commercial payloads were flown or not, so as long as commercial customers paid the marginal cost of their missions, it benefited the nation by enabling cheaper commercial launches. Moreover, if Shuttle prices and launch rate were restricted, the real alternative wasn’t flights on Atlas, Titan, or Delta, but contracts lost to Europe’s Ariane family.
Another ability to curry favor came with STS-61-I’s launch on October 15th from the Cape. Challenger had already shown off for cameras with the teacher in space mission earlier in the year, now it would be the “Hollywood orbiter” once again, as it carried famous NBC science journalist Jay Barbree, who had a record of having covered every single crewed space launch in American history from Alan Shepard’s Freedom 7 to his own STS-61I. The mission plan was packed with events to demonstrate the orbiter’s effectiveness. The primary mission was deployment of a communications satellite and recovery of the Long Duration Exposure Facility (LDEF) free-flying experiment after nineteen months (itself a seven month delay from plan). Besides this, Barbree filmed and participated in middeck locker experiments including some photogenic ones featuring spiderwebs spun in zero-g and an EVA in which a spacewalking astronaut looked into the Shuttle’s windows and waved to NBC’s national audience in video relayed via TDRS. While the American people marveled at how routine Shuttle flights seemed, behind the scenes NASA engineers and astronauts worried about the push to accelerate flights for commercial payloads like STS-61L which would end the year on December 11th.[7]
The fifteen launches planned for 1986 had turned into only 12. This was monthly on average and an increase of three from 1985, but still underperformed hopes. While technicians and engineers struggled to keep up the tempo of flight operations, NASA was engaged in running battles with the Department of Commerce to avoid restrictions on payloads directing business to other rockets. Commerce hoped this would stimulate other American launches instead of the “NASA-subsidized” Shuttle but NASA promised it would only send payloads to the European-subsidized Ariane. Shuttle commercial payloads might not pay their full share of the annual Shuttle operating costs, but at $90 million they were paying the full cost of all parts and labor directly chargeable to their missions. This left NASA in a strange monopoly position: the cheapest per-flight American commercial supplier of launches, but also a key “anchor-customer” and broker for many types of services. While NASA was beginning to find success in addressing commercial markets with Shuttle, they were still their own primary customer. Shuttle was key to everything from space probes to space telescopes to space planning. Satisfying both internal and external customers in coming years would require continuing to build on the routine access to space demonstrated in 1986.
[1]”Space Trek: The Endless Migration” by Jerome Clayton Glenn & George S. Robinson, Page 13
[2] Solid Rocket Booster (SRB) - Evolution and Lessons Learned During the Shuttle Program
[3] Significant elements of this section are derived from the Report to the President By the Presidential Commission On the Space Shuttle Challenger Accident often referred to as “The Rogers Report.” The primary change is that the gust of wind that historically forced the plug to come out, simply doesn’t happen.
[4] Thanks to the article “‘Up Against a Wall’: What 1986 Might Have Been” by Ben Evens for AmericaSpace, 2015-02-08
[5] Masses for the Galileo mission totals 55,572 lbm, while Ulysses totals 53,021 lbm. Jenkins, Dennis R. Space Shuttle: Developing an Icon - 1972-2013 (Forest Lake, MN: Specialty Press, 2016) Vol II, 341
[6] Many may recognize this as the same failure that historically occurred to OV-102 Columbia on STS-93. For writing this section, we pulled from a number of sources including: Spacefacts.de Spaceflight mission report : STS-93, STS-93: We don’t need any more of these by Wayne Hale (2014-10-26), and ’STS-93 at Twenty Years: “A very long eight and a half minutes “‘ by Philip Sloss for NasaSpaceFlight.com (2019-07-22), & STS 93 Ascent Issues "We don't need any more of these" - MCC Loop (Youtube/NasaSpaceFlight)
[7] The Association of Schools of Journalism and mass communication journalist-in-space project, NASA CR-176961, His place with space, By Curtis Krueger for St. Petersburg Times, 2006-09-04
What a great start! Kept me on the edge of my seat and I'm very much looking forward to what comes next!
This short section was one of my favourite parts, how it feels like it's tempting fate? it's all very tense throughout but this gives a feeling of the outside world thinking everything's hunky dory and it's not, and something bigger is coming?
This short section was one of my favourite parts, how it feels like it's tempting fate? it's all very tense throughout but this gives a feeling of the outside world thinking everything's hunky dory and it's not, and something bigger is coming?
The fact it goes "wow only 12 that sucks" sure is interesting compared to IRL, and has quite the implications for where we're going next...can barely wait the next 4.5 hours 
.. The showing of how hurriedness and the rapidly forming cracks in the program are also a nice touch, wonder what will happen there?
.. The showing of how hurriedness and the rapidly forming cracks in the program are also a nice touch, wonder what will happen there?
Post 2: Divergence
Divergence
The opening of the 1980s had put to rest the question of whether NASA could build and fly a reusable vehicle. The new Space Shuttle debuted with its first flights in 1981, and had been certified as operational with STS-9 in 1983. However, the end of the 1980s would determine if NASA’s hopes for the Shuttle would be realized. To sell enough payloads to fill the Space Shuttle at hoped-for cadences, NASA would have to secure the confidence of launch customers ranging from the Department of Defense to commercial satellite providers, while also ensuring they could manage enough launches to meet the schedules. In planning for the tail end of the decade, NASA would need to balance the challenges of continued increases in launch rate, and allocate the system’s capabilities between satisfying commercial customers and their own internal programs like exploration missions. Besides this, on Earth, the agency faced the ongoing fight to receive Congressional approval for full development of Reagan’s Space Station Freedom. That program, in progress since early in Reagan’s first term, now teetered on the edge of slipping out of his administration’s hands and into those of his successors.
Alongside other running battles like the fight to finalize and approve funding for Space Station Freedom, NASA would face pressure from Congress and the Department of Commerce to support challenges to its own dominant position in spaceflight. The largest was the 1987 consideration for Congress to direct funding via NASA for part or all of the launch cost of the proposed “Industrial Space Facility.” This was a single-module, single-launch miniature space station which could receive visits by Space Shuttle and hopefully attract both NASA and commercial interest in long-duration scientific and manufacturing research on orbit. NASA followed directions and put together proposals, but their lack of strong support made it clear that they viewed it as imperiling the more expensive and capable Space Station Freedom. This was still struggling to achieve a final budget (and thus configuration), and NASA feared the existence of another American space station, no matter how small and infrequently visited, could kill the entire Freedom program.[1]
Still, as NASA put together the manifest for 1987, other commercial concepts would advance. Beyond the regular stream of communications satellites, there were also orders placed for two major Shuttle-enabled commercial payloads. Each would both augment and challenge NASA, while relying on NASA for both transport and as a primary customer: the Fairchild Leasecraft and Spacehab, Incorporated’s eponymous pressurized modules.
SpaceHab’s concept was the most straight-forward. NASA had found strong interest in mid-deck locker experiment options. Even with the rising flight rate, more experiments were interested in flying than NASA could fit on the actual mid-deck. Scientific flights like SpaceLab could accommodate these experiments in the Spacelab module’s racks, but there were only so many Spacelab flights available in rotation, and the module’s length left little room for commercial payload deployment or other missions. Spacehab founder Robert “Bob” Citron aimed to fix this, with financial backing initially by Walter Kistler and others. The new company proposed to privately fund the development of a short research module which could fly on missions with other purposes, adding many more middeck locker locations to flights that were already serving scientific or commercial purposes. They would then lease this module to NASA on a per-flight basis to make up the revenue to pay for its development. Other variants could be double-length research or logistics modules to augment the European SpaceLab and resupply future space stations. Canadian firm MacDonald, Dettwiler and Associates was contracted to build the modules, with first launch near the end of the 1980s. This was but the first step of Citron and Kistler’s ambitious vision for private spaceflight. In their plans, they would someday broker space services, provide access to space for passengers and cargo first with Shuttle and maybe eventually their own vehicle, and finally build a private commercial space station for research and tourism possibly by the turn of the millennium. It was but one example of the ambitious visions of commercial spaceflight the Space Shuttle was enabling.
Another vision of the Space Shuttle’s possible contributions to commercial space operations was under development at Fairchild with their Leasecraft. Even as the space shuttle’s flight rate climbed and opportunities for scientific payloads increased, there were some experiments which would benefit from a longer time in orbit than the week to ten days of a SpaceLab flight. Recognizing this possible demand, combined with the Space Shuttle’s ability to recover and return payload, Fairchild signed agreements with NASA and other partners in 1982 to develop a satellite to orbit that would act as the base for what was referred to in the press of the time as “factories in space.” The satellites would provide power, communications, and control functions to their payloads using the same building blocks that had been in use since the late 1970s on Fairchild’s Multi-Mission Modular Spacecraft (MMS) bus that was the basis for the Solar Maximum Mission, Extreme Ultraviolet Explorer, and Landsat 4 / 5 Satellites. The program encountered setbacks, including the departure in 1985 of McDonnell Douglas who had been a primary contractor. By late 1986, Fairchild had made significant advances and secured a new European partner Alenia Spazio, who would provide the experiment carriers based on the SpaceLab pressurized segments.This gave enough confidence to begin booking shuttle missions and selling space in the first platform design for space based manufacturing. Unfortunately, due to issues with the construction of Leasecraft and finding customers willing to be among the first to fly, the debut of the ‘Space Factory’ was delayed first until 1988 and then late 1989.[2]
While NASA sold the future, they also had to keep up with the present. The manifest for 1987 carried 20 missions, including three slipped from 1986. In the end, delays like those which had haunted 1986 compounded. Despite three launches from Vandenberg (for STS-62B, STS-72A, and STS-72B) and a lucky thirteen launches from the Cape, the program still fell short at 16 flights. Though it was a larger increase from 1986 than 1986 had seen from 1985, it was still an average of just four flights per year per orbiter. The lack of spares and labor issues continued to dog the program throughout 1987 despite the slow influx of spares and technicians learning on the job. The situation was improving, but had a long road to fully meeting the demands of the fleet. The active orbiter fleet was in need of longer pauses for heavy overhauls as they accumulated flights and NASA reviewed flight data. In mid-1987, NASA proposed a radical solution. In the program’s early years, a complete set of flight spares for major assemblies had been ordered. Now, NASA proposed to help the active orbiter fleet by completing this as a new vehicle, OV-105, as well as contracting for additional spares for commonly-swapped line replaceable assemblies like landing gear, body flaps, payload bay doors, and elevons. Congress and the White House pushed back that it would be cheaper to merely complete those line-replaceable spares from the OV-105 set for bolstering supplies for the active fleet, but NASA countered that only a full orbiter plus additional flight spares could allow standing down any of the existing orbiters for extended maintenance. Reluctantly, the billion-dollar funding for OV-105 made its way into Presidential requests, and then the FY88 budget late in 1987.
Those concerned that NASA was thinking as much like a launch business as a space exploration agency got a worrying new piece of evidence in 1987. On the eve of finally shipping to the launch site late in the year, the Hubble telescope team was reviewing data when a potentially catastrophic mirror imperfection was discovered. The prime contractor, Perkins-Elmer, had ground their mirror with a novel (and incorrectly calibrated) system, and thus had ground the mirror incredibly precisely to a subtly wrong profile. A backup mirror, produced by legacy contractor Kodak, had been ground but not silvered, meaning additional work would be required to make it a replacement. Moreover, with the optical telescope assembly heavily integrated with the spacecraft, it wasn’t as easy to swap out a defective mirror as it might be for the backyard cardboard telescopes Cape Canaveral and Houston’s gift shops were already stocking as souvenirs for the upcoming launch. The launch was indefinitely delayed as NASA debated the course of action. Finally, it was determined that tearing down the spacecraft’s primary optics would be too disruptive, and instead (thanks to the extreme precision of the mistaken profile) it would be better to develop a new “contact lens” corrector system called COSTAR to sit between the primary mirror and the main instruments. With the telescope designed for further servicing on orbit, over time the replacement instruments could incorporate their own built-in corrections. Still, the design and careful double-checking of the new instrument would push the launch back by years. It was a major blow for one of NASA’s primary scientific missions for Shuttle even as NASA debated how to fit more commercial payloads aboard and the always-touchy question of paid space tourists.
While major changes were underway for the Shuttle program, 1988 wouldn’t see their impact yet. It was what was becoming routine: another year of aggressive manifests, delays, and the bare increase of a single flight over 1987 with 17 launches for the year. Line technicians in Florida and flight controllers in Houston alike cursed NASA management, even as those managers plotted sales and schedules to justify the ongoing investment. As much as four billion dollars a year was vanishing into the Space Shuttle program’s hungry maw. Not even the auctioned-off commercial launch revenue, totaling nearly a billion dollars, could cover the gap as NASA tried to argue for an eight or ten billion dollar space station. At two to three payload specialist seats per flight and 17 launches a year, NASA had room to spare to build political cover the best way they knew how. For the Department of State, Space Shuttle launches were an important way to solidify relations with traditional allies like Germany, the United Kingdom, Japan, and Australia as well as reach out to new potential friends large and small from the Gulf States to Communist China. Sometimes these outreaches would be a success, like signatures of new treaties about biofuel imports with Brazil. Sometimes, they would be less so--the flight of a Chinese astronaut in November 1988 was to have been the first of two, but the second flight was deferred and then canceled after the 1989 Tiananmen Square protests were put down by force. Political junkets would follow for several ranking members of the committees which governed NASA’s budget in both the House and the Senate, and for Under-Secretaries of the Department of Commerce and Department of Defense. It was a routine, even boring year, livened up only by the debut of a new Space Shuttle. This wasn’t OV-105 (now-dubbed Endeavour after a national competition). Nor was it the European Hermes project, which finally received funding approval at the end of 1988 as ESA ministers and European government space leaders agreed on a path forward. This one was Soviet.
The Soviet Space Shuttle Buran lifted off from Baikonur on November 14, 1988. The flight was not a shock in the west--the Energia and Buran programs were well known to western intelligence and some scientists, but some of the Shuttle’s capabilities drew attention. No crew were aboard for the demonstration. On autopilot, the orbiter made a landing after 206 minutes and two orbits and finished mere meters from the runway touchdown point even with heavy crosswinds. The success was trumpeted to the world as the Soviet Union matching the achievements of the West, even as the Soviets tried to evaluate what they wanted to use Buran for in active service. Would they shift their payloads from Proton and R-7 to Shuttle, as the Americans had done for Titan, Atlas, and Delta? To achieve flight rates similar to NASA, they would have to … but there was little appetite or spare funding in the fiercely competitive Soviet rocket bureaus for voluntarily phasing out well-proven rockets. Buran’s second mission was to be made by the second orbiter, while Buran herself was fitted out for transportation to the June 1989 Paris Air Show. It would be many months if not a year before the Soviets were ready to fly again, a span in which NASA seemed likely to turn out a dozen or more launches. In the west, NASA pointed out that for all the Soviet propaganda, they had launched 12 total astronauts in 1988 compared to nearly 120 from America. Indeed, despite the shorter duration of each Shuttle flight, the Shuttle program still had nearly as many crew-days in space in 1988 as the Soviet station program. Between diplomatic outreach to allies with international astronauts and civilian flights like politicians, journalists, teachers, and university researchers, NASA flew more non-professional astronauts in 1988 than the Soviets flew astronauts at all.
The months while the West waited for word about the progress of Buran’s second flight would spin past and then turn into years. In 1989, NASA set two major “firsts” of their own. One was that for the first time the flight total dropped year-over-year, as maintenance issues on Challenger placed pressure on the active orbiter fleet. Only 16 flights were made in the year. However, among them was a more promising first: the launch of the Magellan probe to Venus, boosted by the first of the shorter Centaur-G upper stages. This was the debut of the new stage, purchased in a block buy for Department of Defense launches to replace the solid-fuel Interim Upper Stage. With the lighter stage and the benefit of the added 5,500 lbs of performance from the new Filament Wound Composite solid rocket boosters, NASA was able to make the launch profile of the also-lighter probe work without the stresses on the engines which were suspected to have played some role in the close-run STS-61G incident. The launch marked something of a recovery from the issues of the troubled Hubble program, a confirmation of Shuttle’s applications for probes beyond Earth even as Hubble’s team tested integration of the new COSTAR instrument on an accelerated timeline hoping for a 1990 launch.
At the same time, the last year of the 80s and the first few of the 1990s also marked the debut of the commercial potential for Shuttle. In 1989, the first of Fairchild’s Leasecraft missions was launched, with an initial instrument fitout for NASA and agribusiness purposes flown to a polar earth-observation orbit. In 1990, NASA also made the first flight of a commercial pressurized module from Spacehab, Incorporated. The new single module offered a dramatic increase in the number of mid-deck locker payloads which could be flown. Spacehab also brokered with NASA to offer payload specialist positions for some of their clients. This for the first time allowed astronauts to fly aboard Shuttle without direct selection by NASA. The year also saw the spares issues ease as production picked up on Endeavour and spares both from the new-build lines created to complete OV-105 and those converted from Enterprise began to become available on the flight line. For the first time, many of the most commonly needed line-replaceable components could be ready and waiting to be replaced entirely and then serviced to fly again without having to impair a subsequent mission’s timing by pulling a working module from another orbiter. A new facility also came online, the Offline Booster Processing Facility (OBPF) which had a pair of large hangars each akin to a VAB cell where solid rocket boosters could be prepared and stacked onto a Mobile Launch Platform before being rolled over to the VAB completely ready for external tank and orbiter integration. The facility meant that the VAB proper would no longer have to have long shutdowns in activity for solid ordinance stacking, and along with a fourth MLP promised increased tempo once a fifth orbiter entered the mix. Even without OV-105, the benefits were seen in the jump back up to a new record of 18 flights for the year. To cap the year off, Hubble finally made its long-delayed launch on the renumbered STS-101-L. The window it opened onto the skies brought the promise of a new decade in its corrected vision of the heavens.
For the first two years of the 1990s, though, the President and Congress would debate over what the promise of the decade would be for NASA. Many began to echo the critique that the commercial and flight-rate focus of NASA was distorting the agency’s mission. Space Station Freedom was put forward as one solution, an ambitious project which would fully utilize the capabilities Shuttle had demonstrated and which needed only funding to finalize the design and move into production. However, fractious debates within the president’s advisors and congressional aides continued to stymie the massive budgets needed for Freedom as Shuttle continued to suck down $4 billion a year. Others pointed to a need to turn eyes beyond Earth orbit entirely, with a return to the Moon or a journey to Mars. The expense of such an undertaking would be immense, but those supporters argued it would knock NASA out of its complacent focus on the commercial market and to a new “Apollo-esque” goal. The most visible outcome of this was the (relatively-cheap) funding of advanced design and preliminary fabrication for Shuttle-C, a heavy lift vehicle derived from worn Shuttle spares and aged engines, but offering additional payload volume and mass for possible exploration or defense payloads. With dual-manifesting, it could even offer two major commercial or government launches from the same flight, compressing the benefit of two or three Shuttle launches into a single launch slot--a boon for the Cape’s hard pressed operations. Without dedicated payloads, though, the new launcher’s development was distinctly on the back-burner, not expected until 1994 or beyond. With little firm direction, the main focus for Shuttle in practice remained on the packed manifest of scientific flights, defense payloads, and existing commercial launches, continuing to average 18 flights a year for the next few years.
1992 brought one final lasting change to the program. With Endeavour having rolled out of Palmdale and delivered to Kennedy Space Center in the summer ahead of a planned 1993 debut, NASA was reconsidering their mission naming scheme. The letters used for the “fiscal year plus launch site plus mission plan number” scheme only had 26 available both for missions flown and more critically for missions assigned but not flown. Every year, between a quarter and a third of mission plans were typically overhauled to the extent of a reassigned mission number or canceled entirely to have the payloads distributed over other flights. Some of these were Vandenberg flights using a different set of 26 letters (2-A through 2-Z). Still, with four orbiters and 16 to 18 flights per year, NASA had approached the end of the alphabet in assigning mission numbers. 1992 had seen the flight of Atlantis on STS-121T. Adding potentially five or six more flights per year could overrun the end of the alphabet, leading to STS-131AA or similar improvised solutions. Internally, though, NASA had always maintained a separate designation scheme, which still assigned a number per mission plan and not per flight, but which was in simple numerical order. Beginning in 1993, the scheme would now be used externally, meaning that with 136 missions completed, the first mission for 1993 would be STS-174. Much as some said that the origin of the mission code scheme had been to avoid STS-13, some claimed the switch was the attempt to avoid a year of STS-131 and STS-132 flights (with the use of the mission plan sequence to skip past the remaining launches of the 130s by actual flights). Thus, it was STS-186 that saw Columbia lift off on August 14, 1993 from Kennedy Space Center. It was the 149th launch of the program, the 13th flight for the year, and in rare alignment of flight order and mission number was the 13th flight plan issued for the year, such that it would have been STS-131M under the old sequence. The mission was utterly routine, four more GPS satellites for the DoD’s growing and evolving constellation. The descent … was anything but.
Despite the claim by the Soviets that their Buran uncrewed landing was a major advance, in fact on a normal descent, the entire process was automated for Space Shuttle crews to within the last few hundred feet of altitude, and in theory could be flown all the way to touchdown with the crew only hitting buttons to deploy air data probes and landing gear when needed. With four General Purpose Computers (GPCs) running the Primary Avionics Software System and checking each other’s work, and a fifth standing by with the simpler but more robust Backup Flight Software, the job of a Space Shuttle flight crew on a nominal descent from orbit was more supervisory than stick-and-rudder flying. In spite of this, Space Shuttle crews practiced extensively for the contingencies where this might not be true, and experienced Commander Jacob Weisenberg and Pilot Al Dworski along with every controller in Houston were about to have their practice put to the test.
Commander Weisenberg was a veteran astronaut, with 8 years in the program seeing him only one flight away from the ten launches which had been established in new traditions to earn a change from gold astronaut wings to platinum. Unlike the gold wings of an astronaut who had flown to orbit, the platinum wings were a reflection not just of experience, but of near-emeritus status. Beyond ten launches, NASA tended to rotate astronauts to desk positions like training new astronauts and procedures development to clear flight slots for newer pilots to move up to commanders. Dozens of astronauts had platinum wings, but the fall from those with ten flights to those few with fifteen was sharp. Six of Weisenberg’s previous flights were as a right-seat pilot, with three previous missions plus STS-186 as commander. Pilot Dworski was by contrast a near-rookie, with only three flights including STS-186.
Columbia cruised in a descent orbit towards Edwards Air Force Base, as Kennedy’s weather briefing had restricted its availability. Retrofire was minutes behind them, and though their speed was still nearly orbital, the intersection with Earth’s atmosphere would soon rob them of it. Within forty minutes, one way or another, the Shuttle would be on the ground. At entry interface minus five minutes, Weisenberg switched the state of the four computers running PASS to OPS 304, configuring the system for entry and descent. However, instead of a smooth acknowledgement, instead the crew were greeted with a light no crew wanted to see: SET SPLIT.
In the event of a divergence in the four computer solutions, any three GPCs in PASS could out-vote the one remaining. Any condition where just two computers agreed, whether a 2-2 split or 2-1-1, was called a “set split” and required work from the ground to diagnose. With the time relevance of the coming entry interface, the crew by procedure, had just 45 seconds to make the call with ground controllers to stay on PASS with any remaining good computers in agreement, or switch to BFS on the fifth computer. In Houston, a terse and rapid conference broke out between the Data Processing System (DPS) controller, who was responsible for the GPC hardware monitoring, the Guidance and Procedures Officer (GPO) who was responsible for the software and crew procedures, and the Guidance, Navigation, and Controls engineer (GNC) who was responsible for trajectory, the sensor systems that fed the flight control systems, and the overall responsibility for the avionics system.
The ground controllers quickly ascertained the set split was 2-1-1, with GPC1 and GPC4 diverged from each other, and from GPC2 and GPC3. Capcom quickly requested the crew confirm that the two agreeing computers still had the “full string” of access, with the minimum equipment of sensors and attitude measurement systems to allow them to land. With the choice between two computers running PASS, and a single computer running the more limited BFS, the crew and Houston preferred to stay on PASS. Before the 45 seconds were up, the crew were instructed to halt GPC1 and GPC4 by shutting them down, confirming the status on the “barber pole”. They were clear to continue in PASS on GPC2 and GPC3. Four minutes later, Columbia hit the upper atmosphere at 400,000 ft and a velocity of nearly seven and a half kilometers per second.
The entry data looked good…for the moment. Four minutes into the entry, the orbiter initiated closed loop guidance as the deceleration built to a tenth of a g. The spacecraft was enough of an aircraft now for the computers to bother trying to fly it like one. For nine minutes, even though wrapped in the eerie pink-and-red lights of entry plasma, Columbia’s suddenly reduced computers kept her on track while the crew and ground controllers watched. Thanks to TDRS, the days of communications blackouts from entry plasma were gone. While the plasma blocked the Shuttle from the ground, the ground controllers had data the entire time relayed up to geostationary orbit and back. For the moment, the data all looked good as the Shuttle pressed nose-up into the heat of reentry. At 13 minutes and 40 seconds after entry interface, PASS commanded the first “roll reversal,” a series of S-turns designed to burn off velocity without requiring the orbiter to exit its limited safe entry attitude. The Shuttle slowly rolled right to point the drag vector to starboard, then after a few minutes rolled back gently left to mirror the turn to port, then back to the right to pull the orbiter back to its original trajectory.
At entry plus 15 minutes and 51 seconds, the orbiter reached 200,000 ft (about 60 kilometers) and a velocity of 5 kilometers per second. Though the orbiter had bled off more than 30% of its original velocity, it was still traveling at Mach 17 in the rarified air no other type of winged vehicle could imagine flying in. Both PASS and BFS’s trajectory software changed their modes to ENTRY TRAJ 2 to continue the descent…and once again the dreaded SET SPLIT light came on. The two remaining PASS computers, GPC2 and GPC3, had diverged on the mode switch, and a vehicle with five computers was now down to one. GPC5 and BFS took over the load of supporting the crew.
Flying the orbiter on BFS was much more active for the crew than with PASS. BFS didn’t support many of the automatic controls that PASS offered. While the orbiter could still compute the critical flight data, it couldn’t automatically support maneuvers like the roll reversals by tracking delta-azimuth (the change between the orbiter’s velocity vector and the runway heading). Worse, the system reset the trim values for the roll maneuvers to zero. Flying at Mach 15, the pilots had to manually dial in trim for their elevon and rudder mid-entry. Far from the computer-monitoring of a normal entry, Weisenberg and Dworski were now guiding their orbiter with stick-and-rudder flying. BFS’ more limited ability to plan descent trajectory led Houston to activate contingency procedures. NASA contacted LA Center air traffic controllers, requesting departure holds for regional airports including LAX in Los Angeles and ONT in Ontario, California. They also requested for all aircraft more than ten minutes from landing to be put into holding patterns clear of the airspace. If the Shuttle’s energy management left it seriously short of a landing at Edwards, they wanted every diversion option they could find: LAX, Ontario, and even the orbiter’s birthplace of Palmdale.
The next eight minutes passed like hours for the crew on the middeck and those stations unable to assist, but for the pilots and ground controllers responsible for supporting them they flashed by in a blur of seconds. On the ground, the Data Processing System controller watched the outputs from the remaining GPC like a hawk, as any further failures would leave the orbiter completely uncontrollable while the GPO and GNC updated roll reversal plans and prepared to talk Weisenberg through flying the brick of a glider through terminal energy management (TAEM) with minimal support from the software. Twenty-three minutes and thirty seconds after entry, the Shuttle’s speed had fallen to a mere Mach 5. The air data probes were commanded to extend, but running on BFS, the crew had no way to confirm they were fully locked. Only when they saw or didn’t see data in a few minutes at Mach 3.5 did they know for sure they had the data needed to fully fly the orbiter like a plane, not a spacecraft.
At 25 minutes and forty seconds, Columbia was sixty miles out from Edwards and forty miles from overflying Palmdale. With the advice of guidance controllers echoed up by CAPCOM and what support BFS had to offer, the crew began to fly the delicate balancing act of “terminal area energy management.” This involved balancing burning off enough speed and altitude to touch down while retaining enough speed to still make Edward’s runway. LAX and Ontario airports had been released from their holds, but Palmdale (where after rollout each orbiter had first taken to the sky on the Shuttle Carrier Aircraft) was still standing by, just 20 miles short of Edwards and almost directly on the ground track. Three pilots, Jacob Weisenberg, Al Dworski, and Calvin Baxter on CAPCOM, had their heads locked in conference, though, and the orbiter’s trajectory through energy management was nearly as crisp as it could have been on computer control. Palmdale was left behind as the Shuttle overflew at 54,000 ft, the speed dropping subsonic. The final thruster jets were disabled, and the Shuttle Columbia was now firmly a glider.
Three minutes later, with only a minute and a half to cover the remaining seven and a half miles to Edwards, the crew picked up normal landing procedures. With the BFS the only remaining computer support, the heads-up displays that normally supported the landing were out, but the crew were able to look out the window and verify their approach visually with the Precision Approach Path Indicator lights from Edward’s Runway 04. All crew aboard closed the sealing visors of the launch and entry helmets that went with their blue NASA launch and entry coveralls and G-suit. Weisenberg completed the preflare, and turned Columbia one last time to align with the final descent profile. With just under a mile to go, he called to the ground with the understated humor of pilots, “Edwards Tower, NASA102 Heavy, request landing Runway 04, VFR” - aviation speak for a visual-only landing. Dworski barely had time to pull the gear deployment in response to Weisenberg’s indication before the response crackled back, “Columbia, Houston. Edwards tower confirms clear to land VFR. All yours, Fumes.”
Ten seconds later, Columbia cleared the runway threshold, and then a few seconds later the orbiter bounced as the main gear made contact. Nose high, Columbia skidded down the runway at 230 miles per hour, bleeding off the last 1% of her orbital speed. The speed dropped further, and she settled onto her nose gear. Forty five seconds later, flight controllers leaned back in their seats in Houston at the call, “Wheels stop.”[3]
The STS-186 crew went into post-flight procedures and preparation for turnaround from what had otherwise been a nominal mission. Indeed, other than a slight excess of speed for landing which Weisenberg and Dworski had deliberately kept as “life insurance” knowing the extra ten thousand feet of lakebed runway available beyond runway 04’s official end the landing profile had been well within the program’s normal landing parameters. The flight crew would eat free at dozens of bars, aviation conferences, and professional society dinners for the rest of their lives presenting the story of landing a Space Shuttle by hand-flying it, even though the ground controllers would do the same telling the story of the work of the computer’s failures and the heroic success of the BFS and remaining computer which had let the flight crew make their way safely home. Was STS-186 a story of computer failure and pilot success? Or was it one of pilot and computer success? The causes of the double set-split would be torn down to improve PASS, and BFS was reprogrammed to support the Heads-Up Displays, some of the indications of air data probe deployment, and improve the limited support it could give in terminal area energy management.
The issue would put a nearly two month delay in Shuttle flights. The result was that despite the debut of OV-105 Endeavour, the program would only hit twenty launches in 1993, not the goal of twenty-four. It was also a black eye for the program in the eyes of the new presidential administration, as Bill Clinton debated what to scrap and what to save of George Bush’s ambitious space plans. Bush’s ideas of lunar or Mars exploration had been largely dead-on-arrival with Congress. While Shuttle-C was far enough advanced, cheap enough, and of enough interest for both multi-launch commercial and oversized signals intelligence payloads to avoid cancellation, the core question Clinton’s administration faced was the approval, rejection, or rescoping of Space Station Freedom.
Space Station Freedom had been the marquee “next project” for NASA since its initiation almost a decade prior, though one which had never found Congressional favor to proceed beyond technical studies and analysis. The station had struggled through numerous rounds of redesign, descoping, budget adjustment, re-estimation, and further redesign in turn. The initial proposals for a massive space operations center hosting lunar and Mars-bound tugs, servicing satellites, and large commercial and research laboratories were incrementally whittled down to small collections of modules attached to massive solar arrays. Some in Congress even pointed to the modest commercial success of Fairchild’s Leasecraft and Spacehab’s eponymous research modules as proving that any truly existing needs for long duration space research could be better conducted through cheaper commercially-operated crew-tended platforms, served as needed by Shuttle, rather than by a large and expensive permanently inhabited station. Spacehab had proposed derivatives of their modules capable of testing station components like the International Standard Payload Rack and Common Berthing Mechanisms (mounted to the large flat top side of the module) and even teamed up with Fairchild to present the concept of mounting a Spacehab module to a Leasecraft as a small extended-capability orbital research free-flier, offering both pressurized and unpressurized payload support.
The question of the Space Station was saved by the changes to international relations in the wake of the fall of the Soviet Union. Clinton had placed reaching out to the new Russian state as a critical goal, and the opportunity to bind Russian rocket scientists who might otherwise work with rogue states into a new international cooperation was too good to turn down. The cut-down Space Station Freedom, dubbed informally Fred, along with all its existing Japanese and European components would be integrated with the Soviet-era DOS module prepared for the Mir-2 station to create the new “international space station”. The project faced tense trial votes in 1993, but still secured key votes needed to move into production. During its development, NASA would also be approved for the new “Shuttle-Mir” program, which would see Shuttle fly to the existing Russian Mir Space Station. This would see Russian astronauts flying on Shuttle, like so many diplomatic outreach flights before them, then rendezvous and docking tests with Mir before flying crew rotations on the station and logistics resupplies.
The Space Station’s approval put much of its political hurdles behind it. However, success in the field of diplomacy and politics to win the approval to begin construction brought new challenges. NASA would have to work space station support into the existing tight Shuttle cadence, and their plans for Space Station Freedom would need to be adapted to account for the inclusion of Russian modules and the finalized versions of European, Japanese, and other international participation. NASA’s plans had been approved, putting much of the political risk behind them, but NASA eyed numerous programmatic risks ahead as they worked to make those plans into reality.
[1] Logsdon, John M.. Ronald Reagan and the Space Frontier (Palgrave Studies in the History of Science and Technology) (pp. 476-477). Springer International Publishing. Kindle Edition.
[2] NASA in Deal with Fairchild by John Noble Wilford, New York Times September 23, 1983, Section D, Page 5;
J. Deskevich, "Leasecraft: An Innovative Space Vehicle," in IEEE Transactions on Aerospace and Electronic Systems, vol. AES-20, no. 1, pp. 25-37, Jan. 1984, doi: 10.1109/TAES.1984.310490.;
Accessing space: A catalogue of process, equipment and resources for commercial users, NP-118, September 1988
[3] Information about this form of abort is taken from a variety of sources including:
Shuttle Crew Operations Manual USA007587, Rev. A. CPN-1 December 15, 2008 (specifically Sections 5.4, 6.8, 7.3, 7.4 & Appendix C);
JSC-23266 Approach, Landing, and Roll Out Flight Procedures Rev. B, May 2005;
NASA FACTS: Landing the Space Shuttle Orbiter and SpaceShuttleGuide.com’s various pages, but specifically the one on ‘Navigation’
Last edited:
Aha! My hunch that it would be a Challenger survives TL was right! Super excited, it's a fantastically interesting plausible premise in the best possible hands.
Very interesting that Shuttle's success doesn't save the Station from Fred-ification. This is a great premise, I love that you guys found and used Leasecraft.
Part 3: Apogee
Apogee
The concept of NASA replacing a station with a smaller commercially-enabled crew-tended platform had been rejected thanks to the diplomatic benefits of the International Space Station’s Russian cooperation. NASA was still building on the relationships they were establishing with the growing commercial spaceflight industry. In addition to several flights per year to rotate payloads on Fairchild’s original Leasecraft bus and flying Spacehab research flights, the Shuttle-Mir missions would use a new Logistics Double Module contracted from Spacehab. Even NASA’s marquee projects were to draw on commercial spin offs of Shuttle. NASA was considering flying many of the primary modules for the International Space Station ground-integrated into larger assemblies aboard Shuttle-C to save cost, speed assembly, and reduce the impact of station modules on the number of launches available for commercial missions. Without a Shuttle’s orbital maneuvering system, though, this would need an orbital maneuvering vehicle based at the station to act as a tug to bring the large assemblies into berth with the station. NASA contracted with Fairchild to provide a modified Leasecraft bus to serve in this role. Spacehab, for their part, had begun to act increasingly as a broker for flying not just payloads but astronauts for which NASA didn’t want official responsibility.In 1994, Spacehab would buy out two entire Shuttle flights. One was a flight sold in turn to a consortium of government-sponsored corporations from Poland and the new Baltic nations, making them the first astronauts from those countries to fly on the American Space Shuttle. The flight was encouraged, though not formally arranged, by American diplomats. The other, though, was based out of the United States and the four astronauts included not just researchers and corporate representatives, but also two businessmen whose flights were viewed as more pay-for-flight space tourism than the “commercially enabled space research” Spacehab was nominally coordinating. Ultimately, NASA had struggled with it but decided to allow it. They had more seats on Shuttle than they had strict need for, so making those available for civilians was desirable and provided a means to advertise NASA’s mission and value. NASA even did this themselves with the flight of director Ron Howard and actor Tom Hanks to space on a comsat mission in 1993. While they were formally aboard to film a documentary about flying on Shuttle, it was also to help give them a taste for the realities of spaceflight and microgravity before they began principal photography for the NASA-lionizing Apollo 13. However, they had little desire to see Shuttle become merely a tool for billionaire’s junkets. In both the ex-Soviet and tourist cases, the intermediate broker of Spacehab provided NASA with plausible deniability of just providing a service for at least a few seats a year. However, some would note that the number of interested parties, even at tens of millions per flight, radically outweighed the number of seats NASA was willing to make available, and pursued their own solutions.
These potential tourism providers weren’t the only ones debating alternatives to NASA’s near-monopoly on space launch in the United States. The Department of Defense was reasonably satisfied with the costs and tempo of Space Shuttle operations. The launch rate also allowed notable flexibility. With 24 flights a year finally being reached in 1994, a high-priority mission using standard hardware like block-purchased common solid kick motors and Centaur injection stages with little mission-specific training could slot into the launch schedule given only a month or two of notice. The signals intelligence and even optical intelligence design teams were very excited by the potential of the relatively cheap Shuttle-C, with its substantive payload mass and volume and similar cost to an reusable Shuttle flight (though without any accompanying crew). However, launching on their own Titan IV rockets came with no requirement to coordinate with NASA, and bundling payloads sized for retired Delta or Atlas rockets together to fill Shuttle missions required compromises on the very flexibility Shuttle was intended to offer. Titan was a poor alternative, but if anything ever did happen to Shuttle, it would be good to have another option and even at low flight rates the block-purchased Titan IV would be depleted by the early 2000s.
Moreover, new possibilities for new vehicles beckoned that were built on the foundation of the Shuttle to fix its limitations. Shuttle’s goals had been flight rates approaching weekly and the cost to add “one more flight” of a few tens of millions. It fell dramatically short of this mark, with only half the anticipated launch rate and the cost to add an additional launch per year stretching to between one hundred and one hundred and fifty million dollars.[1] Several approaches had been considered in research for how to fix this. DARPA, the Department of Defense’s research agency, had been hard at work with funding from the Strategic Defense Initiative Organization. In 1993, McDonnell-Douglas’s DC-X vertical takeoff, vertical landing reusable rocket prototype had made its first flights under DARPA funding. The vehicle was intended to demonstrate the control schemes needed for a single-stage to orbit vehicle with turnaround between flights measured in hours or days, not weeks and months like Shuttle. With its first three flights in 1993, it had already demonstrated rocket-powered stable hover and landing, and its turnaround record launch-to-launch was 19 days, half that of Shuttle, with margin for significant decreases. In 1994, the plan was to demonstrate a single-day turnaround and to fly higher and faster with a full propellant load. At the same time, NASA and the DoD were engaged in the 1993 Access to Space study[2], which stressed the need for new and improved Shuttle-complimentary reusable vehicles. While single-stage options attracted attention in the press and were persistently over-valued for their simplicity, two-stage reusable options were found to have significantly higher design margins for weight growth. Thus, the study recommended that NASA and DARPA fund subscale reusable demonstrators for multi-stage partially or fully reusable vehicle technologies with high-confidence and direct applicability to flight. The Air Force also began a parallel program called National Security Launch Vehicle to convert the results of these demonstrators, anticipated to fly in the late 90s, into full-scale launch vehicles to supplement the Space Shuttle. The promise of Air Force development funding for full-scale derivative vehicles and the market potential for growing commercial interest in small-satellite low-orbit constellations attracted the notice of large legacy contractors. Lockheed and McDonnell-Douglas continued to be interested in putting up their own funding for the demonstrators to ensure being among those selected for full-scale NSLV development and operational contracts.
NASA’s interest in vehicles to steal launches or even totally replace the Space Shuttle might have surprised those only familiar with the intense Reagan-era arguments between NASA, the Air Force, the Department of Commerce, and industry representatives. However, nearly a decade later, the Space Shuttle was a mature system no longer in need of more payloads but in fact struggling to bear the number of payloads that had been already booked. The Shuttle fleet was anticipated to make its 200th flight in 1995. The oldest orbiters were already well into their 50th or 60th mission, though Endeavour as the “baby” of the fleet had barely more than 10. Though inspections validated NASA’s hope that the nominal design life of 100 missions per orbiter could be stretched with proper maintenance to 120 or higher, the fleet was still aging. With development time, NASA believed the opportunity was ripe to begin planning for a replacement of the existing Shuttles to avoid a gap in capabilities. This “Shuttle II” was NASA’s answer to the goals which drove the Air Force’s NSLV program. The main question was if NASA’s desire and funding would stretch to a new fully-reusable clean sheet vehicle, or to more limited modification of the existing structures and mold lines for a second generation of improved but similar Shuttles.
While NASA, the rest of the US government, and even private operators debated the future beyond the Space Shuttle, the orbiters continued to follow through on the promises made of them decades before. In 1995, the Shuttle fleet reached a record 25 launches, continuing to fend off challenges to its commercial dominance like Ariane 4. Among those it launched the replacement for Fairchild’s original Leasecraft which had been orbiting Earth for six years, and brought the original module home for outfitting with additional power systems to support pressurized Spacelab or Spacehab modules. Commercial launches hadn’t been the only goal of Shuttle, though. Discovery visited the Hubble Space Telescope for the first servicing of the tremendously successful instrument and in 1995 the Space Shuttle would also operate with a space station for the first time. Cooperation with the Russian program began slowly, first with a Russian astronaut flying on one of NASA’s scientific missions, then with a rendezvous and close proximity operation with the Mir space station. American astronauts would join the station’s crew for short stays, and then finally, Atlantis made her way to the Russian station. For the first time Shuttle would dock to a permanent orbiting independent pressurized space station. Following the conclusion of the highly successful five days of docked operations, including a “fly-around” photographing Atlantis at the station, the Shuttle undocked and began to prepare to return to earth. This was where issues began.
When commanded to close, the doors on the orbiter rotated closed, but there was no indication of them latching. Without confirmation, it was unclear if the doors were secured closed for entry, just loosely closed against the end of their travel, or even still slightly ajar in ways which could jeopardize the orbiter’s safe return. This prospect had haunted the program since before the launch of STS-1, and every single one of the fleet’s hundreds of flights had carried a solution. An astronaut, Evelyn Brown, suited up carrying a set of tools to cut loose stuck pushrods, manually clamp latches into place, and other contingencies. She then wormed her way to the rear of the bay to manually ensure that every fix shy of duct tape was applied to make sure the issue was resolved. The process was hurried along by the limited time the orbiter’s systems could function on reserve cooling with the radiators on the inside faces of the doors retracted. After Brown made her way back to the front of the bay and returned through the airlock, Atlantis could finally descend and land.[3]
As if the shuttles had chattered in their hangars about the concerns over doors being insufficiently secured, the launch of Columbia only two weeks later experienced the opposite issue. When the shuttle first reached orbit, the doors were commanded open. However, the indicators which would show full retraction of the latches in readiness for opening the doors still showed the doors firmly sealed. With only a few hours before the mission’s successful launch would have to turn into an ignominious early abort and return, with the commercial satellites aboard still undeployed, the crew worked to debug the issue. Cycling the latches several times produced no change in the indications. To astronauts peering aft into the payload bay from the control deck, it looked visually like the latches might have moved. A zoom camera lens was employed to verify, and after reassuring themselves and the ground that the latches were open no matter what the indications said, the crew cycled the main bay doors open. The latches were, indeed, retracted and the doors opened smoothly without warping or deforming after a very tense half hour, with more than three hours of margin left to use. When the doors closed and latched correctly at the end of the mission, the entire flight crew from commander Al “Eagle” Dworski to rookie pilot Mark “Forger” Stucky breathed sighs of relief.
The so-called “fortnight of doors” would go down as an unusual coincidence, but another sign of the aging fleet even as they fulfilled their goals of station support. In 1996, Shuttle carried the Docking Module intended to be used with the still-hangared Buran to be attached to Mir, followed by several flights carrying the new Spacehab Double Logistics Module among a total of 24 launches. The launch rate of the Space Shuttles was widely viewed as a success within NASA, but it frustrated some outside of the program. Iridium, one of the low-orbital satellite constellations beginning to approach launch, needed the equivalent of eight full Space Shuttle launches, or two complete Shuttle-C launches, to deliver their full constellation of telephone relays. [4] However, demand for Shuttle flights exceeded supply, and so finding room in the manifest for so many additional payloads in a short window to allow the service to get up and operating was challenging. The debut of commercialized Shuttle-C “Double-Centaur'' launches carrying two Centaur-G stages and two geostationary orbit payloads on the same flight opened up new capacity for other missions, but not enough for all of Iridium’s demands. Iridium couldn’t use the capability themselves thanks to the need to fill out satellites in several different orbital planes, which limited how many could fly on one launch. While Shuttle would fly several times in 1997 and 1998 for Iridium, they would also be among the first wave of customers seeking out alternatives from existing competition like Ariane, but also Russian, Chinese, and even Indian launches - not because Shuttle’s offerings were uncompetitive, but simply because Shuttle’s offerings were unavailable.
Overhauls and maintenance contributed to the persistent struggles breaking free of the 24-flight plateau, as the arrival of Endeavour had allowed each orbiter in turn to stand down for a full year, which saw the equivalent of an airline’s “D-check”. The orbiters were stripped of major systems and torn down extensively for inspection and overhaul, then rebuilt with the latest systems for improved operations and consistency across the fleet. These included the Extended Duration Orbiter system which allowed more endurance through pallets of supplemental reactant tanks, interconnection between the forward and aft RCS pods, and modifications to the Centaur support and venting systems. It also included the installation of new software and hardware for the vehicle’s computers including the ability for the Backup Flight Software to plot more of descent and drive the orbiter’s Heads-Up displays. Little by little, the Shuttle program learned from its mistakes and the aging of the orbiters. The fleet which completed 24 flights in 1996 and 26 flights in 1997 was significantly improved from the orbiter which had rolled out in 1976, made stronger by 255 flights of lessons learned.
These lessons would be put to the test by the preparations to launch the International Space Station. On average, the program would require a launch every two to three months to keep up with the planned tempo of module deliveries to the growing station. This would be roughly 4 to 6 flights a year of capacity which would have to be found within the busy Shuttle and Shuttle-C manifest. Some capacity could come through continued bundling of larger satellite payloads onto Shuttle-C “Dual Centaur” flights. Additional capacity could come from replacing existing NASA flights of Spacelab, Spacehab, and other experiments which could be relocated to the station or simply put off for a while. Some improvements in maintainability could potentially contribute to adding further missions per orbiter per year, like the Block II engines whose improved turbopump designs eliminated the need for removal of the engines from the orbiter between missions. However, other challenges like removing and replacing the hypergolic OMS pods would still limit turnaround and reduce the launch gain benefits. The challenge would simply have to be navigated, and some commercial payloads would simply have to be put off, helping the case both for improved Shuttle-II, for the DoD’s reusable NSLV program, and even for commercial efforts.
The first two launches for the new station would come in the course of a fortnight in late 1998. The Zarya FGB module would launch first on a Proton from Russia, and then the “US Integrated Core” would fly second on a Shuttle-C from Florida two weeks later. The Integrated Core was the name for the assembly of the Unity Node, Destiny Lab, and two Pressurized Mating Adaptors which would be maneuvered to the station by the Fairchild-built, Leasecraft-derived “Orbital Maneuver Vehicle” docked to PMA-2. Once the OMV had brought the combined Integrated Core to the assembly site, Zarya became the active vehicle to dock to the larger assembled stock, linking the first three major modules for the station. The OMV, intended to bring large Shuttle-C launched assemblies to the station and serve as a “retriever” and basis for free-flying pressurized experiments, also was the primary power source for the US side of the station until the solar array trusses could be launched, with 20 kW of solar power providing an average of approximately 8 kW for the station’s systems. With its power, communications, and engines, the Leasecraft OMV would end up proving unexpectedly critical as the Russian Soviet-era Zvezda service module’s launch was delayed more than a year and a half from its original date. In the interim, the OMV could act as a substitute control module, allowing station construction to continue on the so-called “US side of the station.”
In January 1999, the station’s third launch came with the first visit by Space Shuttle Challenger to deliver the third PMA, providing a redundant docking port for the US side of the station, and the Z1 truss with its control moment gyroscopes. Docking Challenger to PMA-2 to allow it to unload and install PMA-3 required the OMV to undock just before the Shuttle entered the station’s 200 meter “Keep Out Sphere,” clearing the forward port on the station which the OV had occupied since launch. Cameras on Challenger would capture this on approach, while OMV’s own cameras would be used for a “fly-around” capturing images of Challenger docked at the station. The OMV then loitered in nearby orbit for several days while PMA-3 and Z-1 were installed and the orbiter could use its arm to install the OMV at its new home on a low-impact berthing port on the Z-1 truss.
The payload capacity of Shuttle-C would be less beneficial to the International Space Station than might be expected. Many modules had been originally designed for the totally-Shuttle-launched Space Station Freedom, and thus fit poorly into multi-module assemblies like the US Integrated Core. Even for lightweight modules like Z-1 or PMA-3, the limited diameter of Shuttle-C’s fairing prevented any modules from being flown “off axis” attached to other modules, and the 82-foot overall fairing length (though longer than Space Shuttle’s) was still not quite enough to take two of the longer inner truss segments. The outer truss sections, P5/6 and S5/6, were to be launched by Shuttle-C, but would need to wait until the OMV could be relieved of duty as interim control module and returned to Earth for servicing and relaunch. Thus, the burden of ISS launches would fall largely on the Space Shuttle.[5]
As American assembly picked up, with the two inboard trusses carrying radiators, the station’s US airlock, and finally the first of the station’s four large solar arrays, the lag in the Russian Zvezda service module became more and more apparent. The last year of the twentieth century turned into the first year of the new Millenium, and while the US section of the station was now power-positive even without Leasecraft and was able to support experiments in its Destiny lab module between Shuttle visits, the Russian portion of the station was no larger than it had been at launch. Zvezda wouldn’t fly until July of 2000, nearly a year and a half late. Until then, the Leasecraft OMV would remain berthed at Z-1 to act as the interim control module. Once Zvezda launched, the last of the inner solar arrays flew for the US side. The tempo of launches to the US side slowed, waiting for Shuttle-C launches and the preparation of the large P5/6 and S5/6 truss segments and the complex international labs, but by the end of 2000 the station was able to take its first crew expeditions, though initially only three crew at a time.
A large portion of the delay in Zvezda was fundamentally financial. While NASA was bankrolling portions of Russian station development, it wasn’t funding everything, and the Russian funding was as lackluster as their overall economy. To make the most of their legacy, the Russians needed to take every chance offered for hard currency, and 1999 would see the birth of one of the most remarkable: BuranCorp. This commercial venture to operate the Soviet-era Buran Shuttle as a private launch vehicle for both payloads and space tourism emerged from those frustrated that even options like Spacehab tourism flights needed to be approved by NASA. Passenger lists had to pass NASA veto authority, and passengers still had to go through several weeks or months of preparatory training to meet Spacehab’s agreements with NASA. Additionally, insufficient Shuttle tickets were available to meet demand, even as many satellite launch missions carried two or three mission specialists and one or two NASA-selected payload specialists who seemed to do little other than justify NASA’s incredibly large astronaut corps and assure every astronaut their annual-on-average flights on the path to their platinum wings and a fat government pension.
BuranCorp promised to combine Western entrepreneurship with Russia’s lower labor costs and Buran’s improved, second-generation all-liquid Shuttle design. Thus, BuranCorp would offer superior and more flexible service for lower costs than Shuttle, including much more flexible (and even exotic) in-flight experiences, and still offer more capacity to meet the rising surge in Shuttle-launched commercial flights which the American Shuttles were insufficient to meet. BuranCorp signed paperwork in 1999 taking formal delivery of the existing flown Buran orbiter and the partially-completed second unit, in exchange for deals for exclusive operation with Roscosmos and Energia. They even signed memorandums of understanding on flights to the existing Mir station on Buran or (if needed) Soyuz to help establish early revenue and interest from venture capital. With this paperwork secured, BuranCorp handed over nearly $50m in cash which went to clean and maintain the buildings for Buran in Baikonur. Meanwhile, the company founders took the resulting images on the road to pitch the “Shuttle that Came in from the Cold” to investors seeking an inroad to NASA’s spaceflight monopoly in the hopes of raising the several hundred million more which would be needed to prepare Buran for its second-ever launch.
While BuranCorp continued attempting to raise funds for the system’s second launch, the United States was preparing its entire second generation of reusable vehicles. After extensive debate, by 1998 NASA’s most ambitious plans for a clean-sheet Shuttle II were largely defunct. The next generation NASA Shuttle would be an evolutionary improvement on the existing vehicles, not a revolutionary break. Over 1997 and 1998, the program’s budget had been debated by Congress as NASA advocated for as much new development as they could get away with. The final version as approved in 1999 would see the basic configuration and aerodynamic design remain, with modifications to the orbiter and its systems to improve maintainability and turn-around, like an ethanol-LOX OMS, RCS, and auxiliary power system. This new system would be non-toxic and capable of being partially serviced on the vehicle, as well as allowing flight consumables to be supplemented with LOX and LH2 scavenged from External Tank residuals before separation. The structure would be revised, stressed for up to 75,000 lbs of payload instead of 65,000 lbs, but with lighter overall structural weight thanks to substitution of new modern composites. The most striking visual change would come from moving the rudder from its original position on a vertical tail to two new vertical stabilizers at the wing tips, enhancing yaw control during high-angle-of-attack descent. The other major change would come with new kerosene-oxygen liquid flyback boosters. This pair of new boosters would separate and fly back to the Cape with jets and wings. This would enable servicing to occur faster and in many cases without ever leaving the Cape, much like the Shuttle, reducing per-flight cost and enhancing processing throughput for launches. The award of the Liquid Flyback Booster (LFBB) would go to Lockheed, widely viewed as compensation for the results of the other major US government reusable launch contract.
Even before NASA’s Shuttle-II downselect, Department of Defense and USAF were finalizing the award of the contract for the National Security Space Launch Vehicle. The final design, coming from Boeing, was a combination of designs from companies which Boeing had absorbed over the competition. To what degree the designs were inherited in the merger and to what degree corporate espionage had perhaps contributed to decisions about merger strategy would be hotly debated. The design paired an SSME-powered winged flyback first stage, effectively a scaled-up version of ex-Rockwell’s X-33 demonstrator even then being prepared for first flights, with a ballistic-but-aerodynamic vertical-landing upper stage, drawing on McDonnell-Douglas’ work on the DC-X, DC-XA, and their privately-funded DC-Y design and testing work before their merger with Boeing. The so-called “Phantom Express” would have a payload to LEO of 30 tons, and was anticipated to debut before the end of the 2000s. Lockheed’s discarded design was a derivative of their failed X-33 proposal, which in turn drew heavily on the drop-tank equipped Starclipper launch vehicle they had studied in the 1960s predating even the formal Space Shuttle program. Instead and largely on their own dime, they worked to add a cheaply-developed expendable second stage option to fly atop a single Shuttle-II LFBB to launch payloads. This would give them their own commercial launcher for the developing LEO comsat market, and the Department of Defense approved of having a secondary backup if possible for commercial competition of launch bids.
The stage was now set for the second generation of reusable vehicles. NASA and the Department of Defense both aimed to build on the experience of the Space Shuttle, combined with modern advances in technology since the original Space Shuttle design had been frozen in the mid-1970s. However, it would be the better part of a decade before any of the newly approved vehicles would see flight beyond subscale demonstrators and component testing. Until that day arrived, the Space Shuttle would have to take up its familiar place shouldering the load of the entirety of American spaceflight.
[1] This is derived from the historical Shuttle Zero Base Study, which evaluated the marginal costs for shuttle flights in the post-Challenger environment.
[2] While the historic study will be different, the drive with a new administration that led to it will still be there.
[3]”Oops” by Wayne Hale, 2019-09-25,
Shuttle Problem Nearly Forces Emergency Landing, AP via LA Times, 1996-03-31, & Space Program Operations Contract Mechanical Systems Training Manual MECH SYS 21002, November 11, 2008; USA006021 Rev. B
[4] Gunter’s Space Page entry on Iridium (retrieved 2024-03-19)
[5] For an example of how NASA evaluated using Shuttle-C to help simplify Space Station Freedom construction prior to the 1993 redesign, this study is a useful guide Shuttle-C utilization for assembly of the rephased Freedom configuration,TM-101658, 1989-08-01
Love this so far! However, I do wonder how much longer NASA can buck the OTL historical failure rate of ~1 loss-of-vehicle per 100 flights. I also wonder if the people behind BuranCorp got a good look at the state of the hardware - and it's support and manufacturing infrastructure - before they parted with their cash...
Part 4: Roll Out
Roll Out
The turn of the millennium found a generational turnover in progress in reusable vehicles. NASA’s Space Shuttle carried on its familiar roles launching commercial and governmental payloads to orbit and tested its new role as a space station construction and logistical support vehicle, but work was in progress to augment and replace it. Subscale demonstrators for the next generation of launch vehicles were beginning testing under both NASA and DoD authority, ranging from the X-33 hydrogen winged booster demonstrator (originally from Rockwell and now absorbed within Boeing) and the X-34 which was similar but with a kerosene engine (originally a product of Orbital Sciences and now absorbed within Lockheed Martin after an acquisition). This testing of subscale vehicles over the next few years would inform the finalization of the recently approved full scale derivatives, NASA’s new Shuttle II with its Lockheed-built “Starclipper” kerosene flyback boosters and the Air Force’s new Boeing-built “Phantom Express” two-stage National Security Launch Vehicle.The decision on funding the next generation of reusable vehicles in the United States and the beginning of subscale testing was roughly contemporaneous with the culmination of Europe’s own plans for a mini-Shuttle. The debut of its launch vehicle, the expendable Ariane 5, was best described as “troubled” when the first launch in 1996 failed, and the second launch suffered from a roll control failure of the core stage which resulted in sub-standard orbital insertion by the upper stage. Flights continued with extensive analysis between them annually until 2000, which saw four launches including the uncrewed two-orbit debut of the Hermes mini-shuttle Jules Verne on mission H01 in September. In July 2001, all was finally set for the H02 demonstration, the first flight of Hermes with crew in space. The flight was planned to last several days, demonstrating control of the orbiter before future flights to the International Space Station. Unfortunately, the launcher’s second stage suffered a major engine underperformance, leaving the orbiter nearly a hundred and fifty kilometers lower than the intended orbit. The orbit was partially stabilized by the crew using Hermes’ OMS, but the flight was still cut short to less than a day for safety. These three high-profile issues with Ariane 5 in the first ten launches of the new vehicle would cast doubt on ArianeSpace’s ability to continue to compete with the American Space Shuttle and on the advisability of ESA’s decision to trust their mini-Shuttle to a clean-sheet launcher design.[1]
However, in 2002, both Arianespace and ESA would be able to put the issues behind them. ArianeSpace managed five flawless launches, and ESA was able to fly Hermes to the space station. On mission H03 in June 2002, the Europeans became the first non-superpower to dock with a space station. After several demonstrations of rendezvous and proximity operations, the first Hermes orbiter, Jules Verne, docked to the forward port of the new Italian-built, American-integrated Harmony Node and the two-person crew spent five days working with the station crew and robotic systems. Later in the year, Hermes demonstrated its logistics chops with a four-week stay at the station on mission H04. The stay this time would be at the aft APAS port on Zvezda, leaving the forward PMA-2 port available for the week-long visit of Space Shuttle Columbia, delivering the Columbus module to the station. The station’s Leasecraft OMV once again served as a flying camera to document the visit of the two shuttles to the station simultaneously. After Columbia’s departure, the Hermes crew would stay several more days to begin outfitting the new European lab module, making a point about Europe’s independent ability to access the station if not to launch their own modules in their entirety.
The logistics provided by regular flights of Hermes and Shuttle were sorely needed. With the Russian service module Zvezda finally integrated and the US solar array and truss completed, the station could support its full design crew. The four US-side international crew would need to make temporary accommodations in Node 2 until the Habitat module was launched but with the expanded (and still-expanding) size of the station, the extra hands were badly needed to run the station’s experiments and maintain its systems. Starting with the fourth expedition to the station in late 2001, the launch of four Soyuz vehicles a year had allowed raising the station’s crew to six. With the eight aboard Columbia (including now-Senator Bill Nelson visiting the nearly-complete International Space Station on his second overall flight to space) and the four aboard Hermes, this made for a total of 18 people on-orbit at once. It was a new record for humanity, and marked a milestone as NASA looked to the future beyond the first-generation Space Shuttle. Shuttle II would have a new large station to service, a “little sister” to augment it for station logistics, and commercial launches to fly. Before it could be debuted, however, the Shuttle program’s flawless safety record would be dramatically and irrevocably marred.
On April 17th, 2003, Space Shuttle Endeavour was fueled on the pad as her crew waited for the launch of STS-468, the sixth mission for the year. Given the gap between flight numbers assigned and flights flown, it was actually only the 395th flight of the Space Shuttle program. The mission was to carry a European Multi-Purpose Logistics Module to the International Space Station for a routine supply run and crew transfer. It was planned to be one mission like dozens of others, but it would not reach orbit. The countdown to liftoff at 4:55 PM local time in Florida was smooth, proceeding as normal though the final holds and into the automatic sequence for flight. Six minutes and fifteen seconds before liftoff, the Shuttle’s computers automatically began prestart for the three Auxiliary Power Units (APUs) which provided for the high energy needs of the engine control systems, gimbal hydraulics, aerodynamic surfaces, and more until the lower demands of orbital operation could be taken over by the fuel cells. Each APU burned hydrazine monopropellant from a dedicated tank with a catalyst to turn a turbine and generate hydraulic power. Each APU drove an independent hydraulic loop. The most critical systems like the elevons, rudder, body flap, and ET umbilical retraction actuators were connected to all three loops and could be powered even if any two loops were offline. Second-tier systems like the main engine gimbals drew on two loops, alternated so that one loop out would cause no loss of gimbal control while two loops out would only cut gimbal control from one engine. Tertiary systems, like engine throttle setting, were connected to only one loop apiece.
By 4:51 PM, four minutes and five seconds before liftoff, the Maintenance, Mechanical, Arm, and Crew Systems flight controller in Houston, who was responsible for the APUs and their hydraulic loops, confirmed all three of Endeavour’s APUs were up and running with normal performance and resulting hydraulic pressures in all three loops. With the final round of go/no-go polling completed, Flight Director Nick Alexander gave the approval to proceed with launch. The three SSMEs lit six seconds before liftoff, but bolts stubbornly held the vehicle to the pad until the ignition of the solid rocket boosters. The main engines reached full power and the off-axis thrust bent the entire craft, deflecting the nose of the external tank by nearly two feet. This “twang” movement began to reverse as the shuttle main engines changed their gimbal position, but only fully released when the solid rocket motors started burning and the bolts that held the entire stack to the launch platform were released, allowing the shuttle to leap for the partly-cloudy skies. The initial flight events were nominal through the roll head-down and the throttle-down through maximum aerodynamic pressure. At two minutes and seven seconds, the two filament-wound composite solid boosters burnt out enough to be safely cast loose. For another 51 seconds, the flight remained normal.
As the time passed 4:58 PM, two minutes and fifty-seven seconds into flight, APU 1’s onboard controller detected an “underspeed” condition - an indication that the turbine wasn’t turning as fast as would be expected given the APU’s fuel intake. This was characteristic of the leadup to any number of terrifying possibilities. The APUs were fast-spinning assemblies of turbines and gearboxes, tightly coupled to a burning source of monopropellant fuel. A turbine rubbing on a wall or a bearing giving way could quickly spark a fire, and turn the APU into a bomb. Such an uncontained failure would fill the vulnerable aft compartment of the Space Shuttle with shrapnel and burning fuel. Without any intervention required, the APU controller shut itself down. MMACS noted the action for the flight director in mission control, but confirmed all other hydraulic systems were still functional. With two good APUs, the Shuttle was still allowed to proceed to orbit - if the situation didn’t get worse.
The MMACS controller continued to monitor the situation. Just as the Flight Dynamics Officer called “negative return” at three minutes and fifty-seven seconds, indicating the Shuttle was too far downrange to return to Kennedy Space Center, the sensors on the Shuttle flashed down troubling data to MMACS’ screen and those of her backroom support. All indications were that APU 1 was shut down properly, inert and not spinning. However, the pressure in its dedicated hydrazine tank continued to drop. The monopropellant was leaving the tank, and it had to be going someplace. The reaction to a monopropellant leak inside the orbiter was almost automatic.
“Flight, MMACS. APU 1 tank pressure dropping. Believe we have a leak,” MMACS said.
“EECOM, confirm that. Any sign of fire?” The flight director was monitoring the entire room, but with the APU 1 issue they had been paying particular attention to the electric and hydraulic systems and replied almost instantly.
“EECOM confirms,” came the word from that station. “Nothing yet but at that rate … it’s when not if, Flight.”
These moments were the kind which Flight Directors trained for and the decisions they hoped to never have to make. A fire in the aft compartment would be catastrophic, but an early abort would put the crew into a non-standard return procedure and would effectively cost taxpayers the better part of (by 2003) one hundred and fifty million dollars. Making the right call under that pressure took a cool which was why NASA Flight Directors were a breed all their own. Alexander made the critical call within 4 seconds of the issue being first noticed.
“Bring them down. CAPCOM, tell Endeavour Abort, TAL. Repeat, Abort TAL.”
Suddenly, what had only been MMACS’ and EECOM’s problem snapped the whole room into action as controllers flipped to well-worn contingency procedure pages in operations manuals never tested outside of simulators.
“Endeavour, Houston. Abort Tee Ay El. Abort Tee Ay El,” CAPCOM called up the connection at four minutes and two seconds.
The Public Affairs Officer broke in over the normal simple broadcast of mission control audio to explain for those watching at home. CNN almost always carried launches live as “Breaking news” simply to have something to fill hours, and some radio news networks were doing the same for those interested in listening to the live broadcast on their drive home.
“This is mission control in Houston, we have a fuel leak and fire risk aboard Space Shuttle Endeavour. The crew has been instructed to abort to a trans-atlantic landing site.”
On the actual communications loops, controllers and the Shuttle crew reacted to the potential disaster in progress.
“Houston, Endeavour, understood. Abort Tee Ay El,” Endeavour’s commander replied to CAPCOM.
The trans-Atlantic landing site for Endeavour’s launch was Zaragoza, Spain, offering a 10,000 ft concrete runway and full set of shuttle-compatible navigation gear. It was already dark in Spain, but Endeavour’s systems would have the ground support they might need to land if nothing else went wrong. Normally, TAL was envisioned for mission modes where engines had failed late in the burn leaving the spacecraft with too much energy for return to Florida. Thus, procedures focused on preserving what energy the Shuttle had managed to build up to make it across the lonely Northern Atlantic with any remaining engines. Now an orbiter with three healthy SSMEs was headed into the abort mode.
In spite of having all three engines running, basic TAL procedures still applied. The Shuttle’s computers stayed in ascent mode, though the orbiter’s trajectory and shutoff speeds were altered slightly. The orbiter would burn to a slightly-reduced MECO if possible with whatever engines it had operational. At the same time, the OMS engines and the RCS thrusters all began firing to burn off propellant from the tanks. The tanks weren’t designed to withstand entry and landing with full propellant loads, and the mass of the propellant would cause additional stress on the wings and thermal protection system if retained. The burnoff down to the normal entry reserve levels was visible in the cameras streaming to the ground as plumes coming off of every thruster. The next three minutes crept past as MMACS monitored the APU systems and the BOOSTER and TRAJ officers monitored the progress of the orbiter’s engines towards a safe entry profile. First, the orbiter reached a speed to be able to press to Zaragoza on two engines. Next came the callout for capability for single-engine Zaragoza, even as the three engines continued burning.
Finally, the velocity was enough to make Zaragoza even without additional energy, but to minimize risk, the burn continued all the way to an only slightly-lower-than-normal cutoff velocity. The three main engines fell silent even as the OMS and RCS burnoff continued. The ET separation fired, and the umbilical links to the tank retracted, closing behind their protective doors in keeping with a normal shutdown procedure. What wasn’t in keeping was the orbiter’s trajectory and altitude. Rather than lobbing the external tank into the Indian or Pacific oceans, the orbiter’s path would put it into the Bay of Biscay. Nor was it normal that immediately after MECO and ET separation, the flight software immediately switched to OPS3, major mode 304 - preparation for entry interface. The orbiter was now configured to glide to Zaragoza, but from its burnout point off the coast of Newfoundland it would take almost 35 minutes to re-enter the atmosphere and touch down in Spain.
Initially the only difference from a normal descent was the landing site, which necessitated contacting Flight Control Centers in Brest and Madrid to let them issue Notices to Airmen and Mariners about the risk of debris from the External Tank, which for the first time in the program would re-enter over the Atlantic Ocean. The computers compensated without issues in their software for the final burnoff of the OMS and RCS propellant, and the orbiter oriented for entry interface. The descent through the roll reversals and terminal energy management was perfect. However, the hydrazine levels in APU 1’s fuel tank continued to drop, and finally the call came that the mission control room had been dreading.
“Flight, MMCAS, APU ONE GAS GENERATOR VALVE TEMPERATURE high,” MMACS said. The temperatures in the APUs routinely rose during entry given heat soak from the outside, but a temperature rise at the gas generator valve meant something else was generating heat.
“Affirm,” Flight said. “Anyone else have elevated temperatures?” Controllers checked their data, and one by one acknowledged they were fine. If there was a fire, it was confined to someplace inside the APU 1 system. Endeavour was still doing Mach 2.2 approximately 60 miles shy of Zaragoza. There was nothing to do but notify the crew, watch other sensors, and pray whatever fire was burning didn’t get worse. For five minutes, every eye that wasn’t locked on landing-critical readouts was checking one or more sensors for the aft compartment, looking for any indications of heat, electrical shorts, or other signatures of a worsening fire in progress.
The temperature at the APU gas generator valve stayed elevated all the way to landing, but no worse indications presented themselves as Endeavour’s landing gear kissed Spanish concrete and a combination of parachutes and speedbrakes fought to bring it to a halt. In one final complication, the commander and pilot were forced to use toe-brake steering to keep the orbiter on the centerline, as the nose-wheel steering was lost when APU 1 was shut down. The airport’s fire brigade was standing by, and a small char could be seen around APU 1’s external exhaust vents. A discussion ensued on the advisability of attempting to apply water or foam to the fire. Hydrazine fires were effectively impossible to extinguish conventionally through smothering, but diluting one with water, foam, or other agents could bring heat below critical temperatures. On the other hand, there was no raging inferno and those same agents could pose the risk of causing further damage and fire risk through electrical shorts or other problems. Ultimately, Alexander and the on-scene ground incident commander decided to focus on getting the crew to safety and let the fire deal with itself for the moment, unless it flared up. Normal shutdown procedures were abbreviated, and the crew were evacuated from the orbiter within ten minutes of touchdown. By 11:45 PM local time in Spain, the crew were off the orbiter, in time for the evening national news breaks on the East Coast. CNN, which had carried the launch since a few minutes before liftoff, didn’t cut away from live coverage (including a hastily scrambled European reporting team for Zaragoza) until nearly seven in the evening when they finally decided there wasn’t more to say over the sight of Shuttle sitting forlornly on the tarmac, watched over by the fire brigade and NASA representatives.
As the crew evacuated to a distance of 1250 feet, the Commander, Rose L. Sullivan formally turned over responsibility for the orbiter to the local NASA representative. Along with the rest of the crew, Sullivan held a teleconferenced debriefing with Mission Control in Houston, after which they were allowed to contact their families before being evacuated to Rota, Spain for the night while NASA arranged for them to fly home to the US. By midnight, multiple planes with NASA and contractor personnel aboard had left Florida, Houston, Palmdale, and Washington D.C. headed for Zaragoza. By dawn, the orbiter had been made safe to approach, and over the course of the day crews were able to begin to investigate the extent of the fire. It appeared that the issue had been with a leak past the worn valves which were intended to isolate the APU from its fuel supply in the event of a shutdown. A small amount of pooled hydrazine had been ignited by residual heat from entry and had smoldered near the APU. No structural damage was found which would prevent the orbiter from traveling, so one of the Shuttle Carrier Aircraft was dispatched along with portable lifting gear to place the orbiter on top of it.[2]
While the investigations continued, NASA did their best to refocus public attention on Shuttle’s future and its successful past, not its present struggles. The year was, after all, the centennial of flight. Discovery, which was off the flight schedule for its Orbiter Maintenance Downtime Period, had been scheduled to make appearances at a number of North American air shows and public events at airports large enough to receive the Shuttle Carrier Aircraft. The intent was to allow members of the public unable to travel to Florida the chance to get up close and personal with the backbone of American spaceflight. Endeavour’s presence in Europe for the spring suddenly offered a fortuitous chance to make lemons into lemonade and extend the tour. After investigators completed evaluating the damage to the orbiter’s aft compartment, and a team of technicians and engineers certified her safe to travel, NASA officially announced that Endeavour was staying in Europe an extra month to put in an appearance at the Paris Air Show, which was to open June 14th. If NASA’s announcement was exciting, the announcement which followed it days later was electrifying: aiming to drum up interest for their so-far-unfunded return to flight, BuranCorp announced that in partnership with Energia, they would be bringing Buran back to the Paris Air Show as well. Both the Shuttle and Buran had attended before, in 1983 and 1989 respectively, but now through coincidence, two flown orbiters from the two sides of the former Iron Curtain would be displayed side-by-side. Photographers swarmed at the chance, and BuranCorp even found some willing to sign partial commitments to fly aboard Buran if the company could raise the entire cost of a return to flight. With this success and NASA’s existing plans, they made a fateful decision: Buran and the An-225 would follow Endeavour and the Shuttle Carrier Aircraft on their return to the United States to appear at air shows where they could hopefully put the vehicle in front of more American investors.
While Endeavour, Buran, and the hastily summoned third Hermes orbiter Edoardo Amaldi posed for the cameras, NASA was deep in the trenches on Shuttle Return-to-Flight. It would take three months to conclude the accident inquiry, and another two to complete inspections and recertification of APU hardware on the other orbiters. Thus, the Space Shuttle which had routinely launched roughly every two to three weeks for the past decade suddenly went the better part of six months without a flight, and even after the return to flight the tempo of Shuttle operations was still slower for another year. The impact was immediately felt in commercial spaceflight sales. ArianeSpace could not ramp up quickly enough to meet the full demand, but still managed to reach ten Ariane 5 launches booked for 2004 and 2005 when added to their existing European governmental demand and Hermes flights to the International Space Station. The slip of commercial business to Ariane helped reinforce the business case for the Phantom Express NSLV and Lockheed’s Starclipper LFBB, which were in the final phases of manufacturing and development alongside NASA’s refreshed and improved Shuttle-II. Some interest even fell into the arms of those in Russia, China, Japan, and India who had been attempting with minimal success to break into the international commercial launch market, though few launches would end up booked.
While Ariane 5 counted their success and Shuttle’s partially and fully-reusable successors took the measure of the market, the BuranCorp team’s gamble of a US tour was in progress. Their first stop was an attempt to pitch the orbiter to New York financiers through a weekend-long static display at Newark airport. While crowds once again mobbed the vehicle, interest in a flight record nearly a decade and a half in the past was less convincing to financiers looking at a market of brand new alternatives. Even the tourism market saw new competition with the Citron-and-Kistler collaboration of the Spacehab Personnel Launch System (PLS-1), a privately funded reusable capsule aiming for launch aboard a variety of the new reusable vehicles. Discovery and Buran would appear together at the Dayton Airshow’s Centennial of Flight, but limited interest could be found even as the company’s funding vanished into repairs to aging buildings, An-225 flight fees, and the bottomless pockets of graft in post-Soviet Russia. Buran’s tour, intended to wrap up in Los Angeles and the Bay Area in the fall, would come to an early end. The An-225 would make low-pass flyover with Buran of the Oshkosh AirVenture, before appearing in static display in Milwaukee where the An-225 could find a long enough runway for those willing to fly in personal planes or be bussed the better part of an hour from the airshow proper. The money finally ran out, and BuranCorp’s creditors came to call. With staff laid off in the United States, Russia, and Kazakhstan, flying personnel back to their homes of origin was valued more highly than the expense of ferrying Buran back across the ocean. To allow the expensive An-225 to return to profitable work, the bankrupt company’s trustees rented a disused cargo hanger and two cranes to unload the orbiter from the An-225’s back and the flown Russian orbiter began a residence in Wisconsin.
The Space Shuttle and Shuttle-C had finally fought their way back to the pre-2003 flight rate by 2005, once again reaching 25 flights for the system. However, the challenges of converting the Shuttle-I infrastructure for the new and enhanced Shuttle-II would catch up to them. Unable to work around the tempo of pad operations and with two pads to call on, NASA would stand down LC-39A between 2005 and 2008 for conversion. In the meantime, flying from a single pad in the Eastern Range, the Shuttles would only reach 17 flights per year. This time the draw-down was foreseen. Boeing’s Phantom Express launcher had recently debuted. With the debut of Boeing’s Phantom Express NSLV in 2006, ArianeSpace wouldn’t be the only beneficiary of this multi-year lull in Shuttle flight. The new launcher would pick up six flights a year from Shuttle, with Ariane 5 absorbing some of the rest. Finally, in 2009, Shuttle-II would make its debut and begin to ramp up tempo, but NASA wasn’t confident enough to stand down Shuttle-I on LC-39B until 2010. In the final years of the program, the last several Space Shuttle missions would be flown out Vandenberg Air Force Base, covering a gap in those missions which Shuttle-II couldn’t fly with a dogleg out of the Cape. This, along with NSLV launches from the same base, ensured that when SLC-6 joined LC-39A and LC-39B in conversion, there would be minimal commercial or governmental impact from the gap in polar access.
The last Shuttle “Eastern Range” flight to ISS in 2010 would be a bittersweet occasion. Though the station was serviced by a variety of international vehicles and would continue to be serviced by Shuttle-II which had already paid its first call, the original Space Shuttle had justified and built the station. Every major module including the Habitat Module launched in 2004 and the Centrifuge Accommodation Module launched in 2009 had flown on the Space Shuttle or Shuttle-C. Even the 2010 Russian Science Power Platform had flown aboard the Space Shuttle Challenger. Only the Russian lab module, Nauka, was scheduled to fly to the station by itself. Even then, with the discovery of catastrophic problems with the module’s fuel systems, NASA and Roscosmos were debating switching it to be deployed to the station by the Shuttle-II and the Leasecraft OMV to avoid a lengthy free-flight and allow total removal of its independent thruster system. [3] The commercial market for satellites continued to grow with second generation Teledesic and Iridium constellations beginning to procure launch service bids. The market for orbital tourism remained strong, served by Spacelab and other brokers procuring flight opportunities. NASA planned to continue offering the same 40-odd flight opportunities a year they had offered nation states, research partners, and more on Shuttle-II. ESA was debating doing something similar with the (fewer) extra seats available aboard Hermes’s two or three flights a year. This was still not enough to satiate commercial demand for crew and cargo access to and from space. In 2005, Spacehab debuted their own independent crew vehicle for logistics, research, and tourism. It had a contract from NASA as a US-built emergency crew return option and backup for Shuttle logistics, and Spacehab also offered it commercially on free-flights, short missions to the international station, and according to their next plans as a crew vehicle to their own potential station.
With the Shuttle program coming to an end, the orbiters began to reach their final resting places. Challenger was the first to be grounded after cracks were found in her wings during maintenance. To avoid the stress of transport, she would remain at Kennedy Space Center in a new public display at the visitor’s complex. For decades to come, tourists would watch videos before entering her display paying homage to the dream of the space shuttle of routine spaceflight, and Challenger’s “fleet leader” status as the orbiter with more flights than any other. Columbia would be the second to end her career, dispatched to the Smithsonian Museum at Dulles Airport, just outside of Washington D.C. Endeavour rounded off NASA’s internal display sites at Space Center Houston’s Visitor Complex, celebrating her TAL mission, her return to flight, and the testing of new systems like propellant scavenging. Atlantis would go on display at the California Science Center in Los Angeles with a spare first-generation External Tank and solid booster casings, showing her launch configuration for the JPL Mars Exploration Rovers Spirit and Opportunity, including a Centaur transfer stage, aeroshells, and a mockup of one of the long-lasting golf-cart sized rovers. In recognition of a similar connection, Discovery was displayed at the National Museum of the United States Air Force in Dayton, Ohio, acknowledging that she had flown the debut Air Force launches from Vandenberg, and the most Department of Defense missions overall The final true orbiter, Enterprise, was displayed at Seattle’s Museum of Flight on the back of N905NA, her Shuttle Carrier Aircraft partner in the approach and landing tests in 1977. Meanwhile, years of bankruptcy proceedings had left Buran stranded in Milwaukee. Russia demanded its return, but was unwilling to pay for the orbiter or the cost of the return flight. In fact, under the nationalist edicts of Vladimir Putin, they demanded that the Americans offer the return for free and possibly even payment with interest for the “theft”. The situation became a diplomatic muddle while Buran languished in a hangar. Finally, Chicago’s Museum of Science and Industry raised money to pay BuranCorp’s creditors and for Buran to be transported by barge down from Milwaukee for display.
The Shuttle program did not come to a conclusion with the retirement of the first generation of Space Shuttles. The Shuttle-II fleet is anticipated to fly well into the 2040s as the centerpiece of the second and even third generation of American reusable rockets and crew vehicles. The future of American government and commercial spaceflight, from LEO comsats to the International Space Station, owe their start to the legacy of the Space Shuttle. Over 1,200 people made flights on the first-generation shuttles, including more than two hundred and figure NASA astronauts and almost a thousand international representatives and civilian spaceflight participants. In total, the program flew astronauts aboard Shuttle more than 3,400 times and included a wide spread of people, ranging from teachers and journalists to poets and politicians, more than every other crew vehicle combined. [4] For commercial payloads, its flawless launch record and relatively low costs still leave the Space Shuttle the mark to beat for second generation systems. The first generation of shuttles may be gone, but they will not soon be forgotten.
[1] Hermes test plan descriptions adapted to this timeline from plans listed here: van den Abeelen, Luc. Spaceplane HERMES: Europe's Dream of Independent Manned Spaceflight (Springer Praxis Books) (pp. 455-456). Springer International Publishing. Kindle Edition.
Ariane 5 failure for H-O2 borrowed from this historic flight: Ariane 5 falls short by Justin Ray for SPACEFLIGHT NOW, Posted: July 12, 2001 at 2305 GMT
[2]Shuttle Crew Operations Manual USA007587, Rev. A. CPN-1 December 15, 2008;
Space Shuttle Operational Flight Rules Volume A, All Flight PCN-1 November 21, 2002
Space Program Operations Contract Intact Ascent Aborts Workbook 21002, October 10, 2006 USA007151 Rev A;
Independent Orbiter Assessment (IOA): Analysis of the auxiliary power unit, McDonnell-Douglas Astronautics for NASA-CR-185568,
JSC-48029 Rev B - Flight Data File - Flight Maps and Charts - 1991-08-28
[3] ISS configuration mostly matches the 1999 plans (”Space Station: Cost to Operate After Assembly is Uncertain”, August 1999, NSAID-99-177, page 29) US Orbital Segment configuration mostly matches the 1999 plans, with an extra PMA to dock PLS/ACRV. The Russian Orbital Segment will roughly match the historic configuration, with the exception of the Science Power Platform having been launched instead of Rassvet.
[4] Jenkins has astronaut counts for OTL as 355 and 852 seats. Jenkins also notes that the OTL shuttle was responsible for a majority of all payloads placed into orbit, here it would be the vast majority. Jenkins, Dennis R. Space Shuttle: Developing an Icon - 1972-2013 (Forest Lake, MN: Specialty Press, 2016) Volume III-400[/URL]
Awesome TL! The little details like MER being launched by Shuttle and Buran being stuck in Wisconsin are great.
I love this Shuttle-II design, it feels so grounded while still being a significant improvement. Though it doesn't seem like the Shuttle TPS vulnerability was meaningfully addressed here? They got lucky avoiding any loss of crew, but debris from the ET would still likely be an issue with Shuttle-II, which is a bit worrying.
I love this Shuttle-II design, it feels so grounded while still being a significant improvement. Though it doesn't seem like the Shuttle TPS vulnerability was meaningfully addressed here? They got lucky avoiding any loss of crew, but debris from the ET would still likely be an issue with Shuttle-II, which is a bit worrying.