Boldly Going: A History of an American Space Station

I find it amusing how proposals for increasing STS performance start looking more and more like Russia's Energia designs.
'Space is all about physics, and physics don't care about your politics!' - something I remember hearing somewhere

In a general way, no physics don't care about your politics but it could also be said that physics doesn't care if you study it or not :D

The thing was this kind of stuff was studied early on in the evolution of the STS but bringing the engines back was a major design goal so you end up with either a ballistic (landing in the Australian outback or northern Mexico) or a lifting recoverable engine pod but which actually worked 'against' the main, (as seen by MSC :) ) goal of the system which was a requirement to fly manned, every single flight. In fact a lot of the late run STS concepts had the ET mounted on an 'in-line' SRB or LRB to simplify the system but in the end the 'simpler' system won out.

It could have been a bit more radical though:

In seriousness, the best reference I found for ensuring the markings matched the period was this site for paper scale models. Their attention to detail is borderline Trekkie!


That site is amazing. I'm considering making some slight edits in the tracks and using them for my "USS Enterprise," a 300,000 (sic) Ice Crawler on an Earth inspired by "A Pail of Air" if I resume working on that one.

You, get back to work! And also get to work on all the rest of the ideas you have. We've determined that if you fully cut out sleeping, eating, and anything like a 'social' life you should be able to just barely maintain the writing schedule we've planned for you... :)
(I like that idea as I enjoyed "PoA" but didn't really buy the 'ballistic nuclear rocket' transport idea :) )

(I like that idea as I enjoyed "PoA" but didn't really buy the 'ballistic nuclear rocket' transport idea :) )
Ice trek had no nuclear rocket transports; it's (so far) limited to the ground.
I've started rewriting it, but it is, for now, a distant third to The Masquerade and Reach For the Skies.)
So from the comments it would appear that a top and tailed tank would in theory be possible, whether advantageous is a another matter entirely
My main point was that having habit and docking station prebuilt on earth and fitted out is a lot quicker and easier than working in the tank. Also having direct access to the tanks from either end is easier and quicker than going down a tortuously curved inflated access tube, As another bonus it may be possible to build an annular access hatch/ lock that permits the insertion of relatively large items into the tank. If it is practical to open the tank to vacuum (at start of mission say) then with both inner and outer lock doors open there is no limit to the length of structure inserted. I know this might be a load of rubbish but comments would be useful and illuminating
Having a module on the top of the tank is annoying, as that's where the LOX tank boiloff gaseous oxygen vent arm connects, the so-called "beanie cap" :

Using an ACC was thus conceived as a way to get more mass aboard a mission than the Shuttle could fit into its payload bay, given payload densities and the limits of where the Shuttle center of mass could be for a launch abort. In this case, it'd allow a larger rigid node, with easier access, though the intertank passage would be complex and the connection from that ACC node to the orbiter would have to tunnel right through the boat-tail, the complex array of pipes and tanks feeding the Shuttle's 3 SSMEs. Even with that complexity, arguably using ACC on Enterprise might have been a good idea in this timeline if they knew in 1983 they wouldn't have to launch until 1987 or later. Unfortunately, they don't know and they're in a budget and time crunch. ACC alone was supposed to be a three year project, which is most of the time they have available, so they don't end up using it. I'm sure there's studies of an ACC Enterprise or Enterprise-follow-on that some Marshall or Johnson engineers generate inside the Space Station Enterprise Program Office, but it doesn't fly. Without landing requirements, they can put most of what they need into the Shuttle cargo bay. As for the intertank tubes, I'm sure that the benefits of an ACC rigid node compared to the as-flown configuration ITTL would be a point of wistful comparison later in Enterprise's orbital life. For the moment...well, you'll have to wait and see.
Not really. Being able to use SSMEs to close to the target orbit was expected to allow the carriage of a fair amount of weight in both the Orbiter and the ET payload unit. Certainly the Aft Cargo Carrier, probably the most developed concept, envisioned missions carrying payloads in both the orbiter and the ACC. For example, if you read one of Portree's posts on the subject, you'll find that Martin Marietta envisioned

Flight 1, a mission with an initial 160-nautical-mile orbit at 28.5° of inclination, would see three satellites with identical solid-propellant upper stages launched in the ACC: the 8848-pound Brazilsat/Payload Assist Module (PAM)-D, the 8848-pound GOES/PAM-D, and the 9399-pound Telsat/PAM-D. The Orbiter, meanwhile, would carry a 58-foot-long, 14-foot-diameter "large observatory" with a mass of 18,700 pounds.

The advantage would come partially from circumventing center-of-mass limitations, partially from circumventing payload bay volume limitations, and partially because more of the necessary delta-V could come from the efficient SSMEs and less from the inefficient OMS engines. Of course the first isn't an issue on this flight, and neither is the third, but the second is.
Note these totals are only about 45,800 lbs, not accounting for the ACC's own structure. Even cramming the ACC and Shuttle cargo bay, this is less than the nominal capability of the STS stack to this orbit., which ought to be about 60,000 lbs.
So,pretty much Hubble?
Not quite, Hubble ended up shorter and about 6,000 lbs (32%) heavier. Same Large Telescope program, though, just an earlier phase before it put on weight.
Thanks Guys, a Great Time line thank you for the response to my query. I once spent an entire day at the cape including time in the big shuttle training mock up. Great day!
One important thing to remember about the Aft Cargo Carrier, is that the use of it was based on a possible shuttle payload capacity in excess of the 65,000 lbm that the bay was rated for. If the SSMEs were run at 106%, and the steel-case SRBs were replaced with the filament-wound-case units designed for polar launches, the payload to an east-bound orbit out of KSC would have been around 72,000 lbm. This extra capacity could not have been loaded into the cargo bay, and thus could not have been used directly without an extra payload volume. Furthermore outside of large upper stages, it was very hard to weight-out the shuttle bay before it volumed-out. Historically, only the Centaur G' payloads came close.
Part 7: STS-38R launch: a race to activate Enterprise before issues emerge
Boldly Going Part 7

With the successful single launch of OV-101 complete, the flight of Space Station Enterprise seemed to clear years of doubts about the program in a single shot. Cheers erupted in Houston, Florida, and in the homes and offices of hundreds of thousands of people who had worked on the station as the ground controllers verified the station was talking to TDRS, then as that relay carried along confirmation that Enterprise had completed the first OMS burn to stabilize her orbit. Soon, a second signal carried confirmation that the orbiter-turned-station’s computers had activated the payload bay door actuators, and onboard cameras downlinked the welcome image of Enterprise’s doors swinging smoothly open in space for the first time. However, the last event on the launch timeline was the most critical of them all--no one on the ground could relax until the computers completed their first programmed activities and deployed the station’s keep alive solar arrays. Too many still remembered that Skylab’s panels had been the source of its problems, and were bracing themselves for a desperate effort to get off Atlantis’s STS-38R launch if anything failed. As TDRS passed the station from satellite to satellite, the downlink capability occasionally faltered. The moment of solar array deployment found the station passing through a window capable of transmitting only telemetry, which recorded the signals indicating that the two solar array sections facing up and out of the bay should have begun to deploy. The Marshall and Houston teams held their breath in spite of the early indications. Deployment motors being triggered meant little, and nothing was for sure until the panels were seen to be open and providing power.

Finally, the downlink flickered back into high-rate capability, and the video picked back up to capture the station’s panels cleanly extending from the Shuttle’s payload bay. Even knowing that the process of fully deploying the keep alive panels would take another half an hour, Houston’s mission control team erupted into cheers--the most critical aspects of deployment were already behind them. Soon, the panels were extended and locked, and life-giving electrical power was beginning to top off the station’s batteries. Space Station Enterprise had reached orbit successfully and now began her service in space. President George H.W. Bush and Former-President Reagan met with Administrator Richard Truly for photographs and speeches commemorating the success of the program. Former-President Reagan’s speech was particularly memorable, as he marked the Space Station Enterprise program’s success as a “Symbol of American Freedom and Enterprise,” comparing it favorably to the capabilities of the Soviet Mir. President Bush’s remarks were more limited, hailing the success, but his speechwriting team had concentrated their superlatives in space for later that month, when they were preparing to announce a major new initiative in exploration. For the moment, the legacy of Enterprise’s launch was marked more by the former president who had initiated it than by the President who would shape its utilization.

With Enterprise’s solar arrays deployed and all STS-37R ascent activities completed nominally, the need for a rushed launch of Atlantis and STS-38R for a mission to “Save Enterprise” as Pete Conrad had once raced to “Save Skylab” was gone. This came as a relief, as unexpectedly poor weather at the Trans-oceanic Abort Landing (TAL) sites haunted Atlantis during the STS-37R countdown. Several press questions and counterfactual what-ifs have hinged on whether NASA would have risked the launch anyway, given the tiny window during launch where such a transatlantic abort was needed and that Space Station Enterprise’s success might have hung in the balance. Then and now, the official NASA stance was and has remained that the safety lessons of Discovery for risk versus reward were clear. Finally, the weather front passed Spain and Morocco, and the first countdown for STS-38R took place on the morning of July 9th. For less complex missions, NASA had cut backup crew assignments prior to the Discovery disaster. Only mission-specific personnel like Payload Specialists were assigned specific backups. Others roles like mission specialists and pilots were assumed to be able to be pulled at need from the general astronaut pool for any specific mission. However, the Space Station Enterprise activation and checkout mission had enough mission-specific complexities driving increased training requirements for the crew that NASA had scheduled full prime and backup crews even before Discovery’s loss.

When first planned in 1986, John Young had assigned himself as commander of the outfitting crew, continuing his record of seeing off the first mission of each new type of module launched aboard Shuttle, including the original STS-1 flight, the STS-9 mission which debuted Spacelab, and the deployment of the Hubble Space Telescope aboard STS-61-J. However, his outspoken critique of NASA management in the wake of the Discovery disaster lead to Young’s promotion out of the astronaut office, effectively removing him from day-to-day management and more relevantly cutting him from the active flight list. He had already personally recruited Joe Engle as the backup commander, and intervened to persuade Engle (who had been considering retiring from spaceflight after the disaster) to instead make one final flight as prime commander for the mission which would commission Space Station Enterprise. Engle’s record was long: three suborbital X-15 flights over 50 miles, a near-miss with lunar missions aboard Apollo 17 which saw him bumped for geologist Harrison Schmitt, and two previous Space Shuttle commands. Moreover, Engle had even flown Enterprise herself in the approach and landing tests of 1977. With the two-year flight stand-down and shifting assignments following the planned Return to Flight, Engle would have his pick of the astronaut corps for the prime crew for the Space Station Enterprise activation mission.

As a result, the rest of the Atlantis flight crew was similarly experienced, including several who had flown multiple Shuttle missions. Experience with the Spacelab module was also sought, given its contributions to the Leonardo Laboratory Module. Pilot Steven Nagel was on his third spaceflight, having previously flown with Spacelab aboard STS-61-A in 1985. On his first spaceflight, he had also had experience with the Robotic Manipulator System (Canadarm) as a mission specialist aboard STS-51-G. On that flight, he had helped use the arm to deploy and then retrieve the Spartan 1 free-flying astronomy satellite. The senior astronaut in the crew by time in space, however, was the prime crew Mission Specialist 1, Owen Garriott. Garriott’s time in the astronaut corps dated back to Apollo, and his experience with Skylab had led to him being appointed as an astronaut liaison to the Space Station Enterprise Program office, providing recommendations for designing the station’s interior for long-duration space operations and consulting on what could be expected for astronauts outfitting a space station while in orbit on missions measuring not days or weeks, but months. There was no one on the flight list more experienced with Enterprise’s new incarnation, and Garriott had been a natural for Young to recruit for the prime crew in 1986. Garriott had eagerly leapt at the chance to implement the results of his hard work by flying to his second space station and adding more days to his existing record: 59 days from his time on Skylab and 10 days from his flight aboard STS-9 with the Spacelab debut. After Discovery, Garriott had (like Engle) been considering retirement, but John Young’s offer to keep him on the Enterprise deployment prime crew had been irresistible for a man who had thoroughly enjoyed his time aboard Skylab more than a decade before. Garriott delayed his retirement by two years to see the mission completed.

Two more veteran Mission Specialists joined the final STS-38R prime crew. The first was Mission Specialist 2 Norman Thagard, who had already flown on three Shuttle missions. This record, accumulating seventeen days in space, included a Spacelab flight, the Galileo deployment mission STS-61-G, and the complex RMS operations of STS-7. As a licensed physician, Thagard would also provide on-orbit monitoring of the crew’s health during one of the longest and most intense Shuttle flights since the program’s start. Mission Specialist 3 on the STS-38R prime crew ended up being Kathryn Sullivan, who had flown EVA during the Hubble deployment on STS-61-J. While Young had slotted her into the prime crew without certainty of her being able to take on the complexity of the Enterprise deployment flight less than a year after the Hubble launch, she was readily available with the additional two-year stand-down. Sullivan had spent the time reviewing EVA procedures for the deployment of Enterprise cargo bay systems and the confirmation of passivating and sealing the massive oxygen and hydrogen tanks of ET-007 both with Garriott and with the last American on STS-38R. This was Mission Specialist 4, rookie Pierre Thuot, who would complete the Extra-Vehicular and Intra-Vehicular Activity specialist team. The final crew member would be the first international visitor to Space Station Enterprise, Ulf Merbold of Germany representing the European Space Agency as a Payload Specialist, drawing on his previous flight with Spacelab. It was a mark of the critical nature of the program that all but one of the crew had previously flown to space, and Engle had trained his crew rigorously for their purpose.

All told, Atlantis carried up seven crew members on her STS-38R mission. The planned mission duration would be 11 days, one of the longest Shuttle missions to date. In the future, Atlantis would be able to draw on Space Station Enterprise to stretch her orbital endurance, using a new system which would allow Atlantis to draw power from the station. In time, all of the remaining orbiters would receive the modification, but Atlantis’ construction after the program’s approval meant she included the capability from the day of her rollout at Palmdale. The capability would not be used on this flight, at least not as originally intended. Still, the fifth mission for the youngest orbiter in the fleet was in many ways what she in particular and her kind in general had been built for: working with a space station for the servicing and deployment of complex payloads in space.

Engle and the rest of the STS-38R crew made rendezvous with Enterprise in the station’s 39-degree, 350 km orbit on July 11, 1989. They had spent their first full day in orbit, Flight Day One, chasing down the station from their lower elliptical parking orbit. The results greeted them on Flight Day Two, July 11th as they approached their target. It was, in one sense, the first arrival of Shuttle at a space station--one of three the Space Shuttle would visit during its history. In another, it was the first rendezvous of two orbiters in space, and the first docking of the Space Shuttle to anything at all. In Atlantis’s cockpit, what it looked like was a tremendous challenge for the flight crew. The enormous white external tank was starkly visible, and the crew called “tally ho” on the station from kilometers out. The station’s bulk grew slowly. With the distance disguised by the clarity of vacuum, Enterprise and the external tank hung in space like a model, looking at first glance no larger than the small satellites Shuttles had deployed and retrieved in the past. It was only in the final minutes, as the distance melted away, that the true scale of the station became clear.

(Images by @nixonshead)
Last edited:
...when [the Bush administration] were preparing to announce a major new initiative in exploration... of three the Space Shuttle would visit during its history...
Some tasty bits of foreshadowing there, Mr. Pi!
The extra pictures are just the capstone in another great post. Very much looking forward to where Enterprise and the rest of the Shuttle fleet head next!
Just wanted to throw this out there, in a happy coincidence Periscope Films has released their footage of the Enterprise rollout ceremony today.

Though, how bad he turns out to be *here* will depend in large part on what George H.W. Bush intends by his "major new initiative in exploration."

In our history, that initiative was the Space Exploration Initiative. It was stillborn largely because Bush did not understand (until it was too late) that Truly opposed that vision as the wrong path for NASA, and worked with likeminded managers at NASA to undermine it at every turn.

The problem with Dick Truly was not so much that he was a bad administrator in se (though some were of the opinion that he was even that, too), but that he was very much the wrong administrator to execute his president's vision.

But it is entirely plausible that in this timeline, Bush will still pick him for his administrator, so I'm certainly not being critical of our authors here. I think they have done a fine job of generating a timeline that is not only *plausible*, but even quite *probable*, given their point of departure. The logic that drove George Bush to appoint Dick Truly will have a lot of the same force here, too.