Wow this website couldn't have given me a better birthday present
Right Side Up: A History of the Space Transportation System
- Thread starter Polish Eagle
- Start date
Now at just past T-24 hours to entry interface. All systems appear nominal at this time. Running final pre-entry checks.
Yay!!!!
You know, reading this TL really, really makes it clear that the focus of NASA and all the space enthusiasts was so, SO wrong. Reusable spacecraft? Ya, that's nice, but the boosters are the big ticket items.
Why, oh, why, couldn't someone have had a brainstorm and done it your way.
Thank you for this TL. I really appreciate it. (and weep at the lost opportunities)
Chapter 11: Retropropulsion
“Flight controllers here looking very carefully at the situation, obviously a major malfunction……..We have confirmation from the Flight Dynamics Officer that there has been a failure of the upper stage. The crew of the Space Lifter Liberty are continuing to prepare for their return to Earth.” -- Public Affairs Officer Transcript, STS-116 Mission
Chapter 11: Retropropulsion
110 kilometers above the North Atlantic, the Space Lifter Constitution drifted slowly, lazily, its flat underside pointed out at the stars, its rounded dorsal hull and immense rudders down to Earth, while keeping its engines pointed east. As the Lifter reached the peak of its trajectory, the onboard guidance computer calculated the precise orientation and duration of the burn that would be necessary to return Constitution to her launch site. As the seconds ticked by, the peroxide attitude-control thrusters on Constitution’s nose and tail fired softly, gently, keeping the Lifter pointed at the correct azimuth.
Then the center F-1B restarted, together with two of the outboard engines, pushing Young and Crippen back into their seats with over 4 Gs of acceleration, as the Lifter turned itself around, bleeding off the speed it had husbanded through its flight so far and then pushing itself back toward Florida. The cabin shuddered as the remaining propellant rushed through the turbopumps into the combustion chambers, the mass of the booster dropping slowly as kerosene and oxygen were driven out of the tanks at phenomenal rates. 15 seconds after the burn began, there was a sudden jerk--the outboard engines cut off to control acceleration. The center engine kept burning, throttled up, in fact, at a more modest 2.5 Gs, climbing slowly, slowly, until it, too, cut off, with all but the vapors in the propellant tank spent.
For a brief moment, Young and Crippen were back in microgravity, the earth slowly, imperceptibly, growing larger in their windshield. Then the guidance computer moved on to the next step in its algorithm, continuing the electromechanical dance it began at launch. Valves opened in the nose and tail to release hydrogen peroxide onto a catalyst bed, where, in a burst of heat, it dissociated into steam and hot oxygen gas, which blasted out into the near-vacuum of Earth’s thermosphere. The valves opened and closed in unison, to impart a precisely calculated momentum to the Lifter. Seconds later, the great vehicle was again pointing in the direction of its own travel, its trajectory altered, steeper, but with a slower atmospheric entry speed, and a point of impact considerably closer to Florida than it had been. Like exactingly trained ballet dancers, the valves on the opposite side of the vehicle repeated the movements of their counterparts, cancelling every kilogram-meter-squared-per-second that the earlier bursts of gas had imparted, zeroing the Lifter’s rotation and keeping its nose pointed straight down the line of its descent. For now, the great ship yet operated beyond the Karman line, where the laws of aerodynamics give way to a much purer expression of Newton’s Laws. The thin wisps of air curling around the ship’s wings acted more like discrete clumps of atomic oxygen and nitrogen and helium than a continuous fluid. But soon, they would thicken, and their grasp on the ship grow. The coming struggle would force man and machine alike to the very edge of their capabilities.
Despite the frantic efforts of mission controllers to salvage the Magellan probe in the minutes after the failure of the J-2S-2 engine to start, the spacecraft and its upper stage soon reached their apogee and began falling back to Earth. When it became apparent that the spacecraft could not complete its mission, controllers ceased attempts to light the engine belatedly to avoid risk to the returning Lifter booster. After a safe gap was opened by the Lifter’s retro burn, range safety officers remotely triggered the self-destruct mechanisms on the S-IVC, activating a number of shaped-charge explosives that destroyed the liquid hydrogen and liquid oxygen tanks, whose contents swiftly boiled in the near-vacuum of Earth’s thermosphere. The small chunks left of the S-IVC and its payload reentered Earth’s atmosphere and disintegrated further shortly afterward, and were torn into ever-smaller fragments as systems built for the forgiving vacuum of space were subjected to the phenomenal loads of hypersonic flight.
Even before the stage was destroyed, engineers across the United States, at Kennedy Space Center, Johnson Space Center, and Marshall Space Flight Center, and at Rocketdyne’s Canoga Park headquarters in California, were poring over the telemetry the stage beamed back in order to determine exactly what had gone wrong. Over the course of the next several months, this telemetry would be combined with intensive scrutiny of all records of the hardware that had been mounted on STS-116, and a regime of static tests of other S-IVCs to identify exactly what had malfunctioned. Every moment for months predating T-0 to the recovery of washed up fragments of hardware on Florida beaches in the weeks following the failure was collected to piece together the cause of the failure that had happened on April 6, 1988.
While the engineers worked to identify why the S-IVC had failed to complete its mission, NASA and USAF managers began coordinating with their counterparts at Martin-Marietta to bring the mothballed Titan launch pads, LC-40 and LC-41 at Cape Canaveral and SLC-4E at Vandenberg Air Force Base, back online. Though a previous Lifter flight, STS-113, had launched the final KH-12 just a few months before the Magellan failure, satisfying the Department of Defense’s need for the very high-mass optical reconnaissance satellites, the USAF still had a manifest of electronic intelligence and general communications payloads to loft--ones that, depending on how long Lifter remained out-of-service, might have to go up on Titan III. For Martin, deep in planning with the newly-formed Trans-Pacific Launch Industries (TPLI) for their own reusable vehicle, the contracts to activate the Titan III contingencies was a mix of benefits and drawbacks: engineers had to be pulled from design meetings to dust off stages which had lain preserved in warehouses for years.. At the same time, the contract brough important revenue to Martin at a time when TPLI was gearing up for major investments. Thus, even as the Space Lifter stand-down helped reinforce Martin’s lobbying to politicians about the benefits of a second (even partially) American reusable launch vehicle, Martin worked diligently to restore operational status to a rocket many had written off as condemned to the history books.
Space Transportation Corporation’s commercial customers like Geostar, Intelsat, and a host of foreign companies had their own back-ups planned. Though Europe’s Ariane was not nearly as powerful as Lifter, and somewhat more expensive even so, it had a crucial advantage over the Lifter system in 1988: it was operational. While Lifter stood down, Ariane won over half a dozen new payloads--three Intelsat geostationary communications satellites, a Swedish communications satellites, a Japanese communications satellite, a Japanese meteorological satellite, and a British communications satellite named Skynet. Though the increased revenue was a welcome bonus to Arianespace as it worked to bring its new reusable launch vehicle from the drawing board to the runway, no one at Arianespace or the European Space Agency was under any illusion that it would last--sooner or later, they knew, Lifter would be back, and then Europe would be playing catch-up again. Worse, even the newer Ariane 2/3 family was incapable of lifting many of the satellites which had been originally manifested for Lifter. Some commercial customers began to demand that if the stand-down went on beyond some time, STC should work with the US government to make the USAF’s husbanded Titans available for commercial payloads that STC could no longer deliver.
As the commercial launch market adapted to Lifter’s indefinite stand-down, and the US government brought its handfuls of stored Titans out of storage, the initial tension at NASA and STC began to give way to a sense of relief. Though it had been over twenty years since that horrendous winter night, many NASA managers, particularly the older ones who had actually worked for the program in those days, could not shake the specter of Apollo 1. It quickly became clear that the cause of the accident had not been unique to the payload--the failure that splashed Magellan into the western Atlantic could have just as easily put a Shuttle crew into the unenviable position of having to either maneuver around for a Return to Launch Site abort or ditch into the ocean, far from any naval recovery force. Though the Shuttle was rated for a suborbital reentry, and though the crews were all trained in mid-air evacuations, no one wanted to put that training to the test.
As much, then, as this failure vindicated the arguments made by Boeing’s engineers all the way back in 1971, that the Interim Semi-Reusable System was safer than the competing Thrust-Augmented Orbiter (TAOS) model because it did not necessarily need to carry crew all the way to orbit, it put the entire Space Transportation System under an uncomfortable congressional microscope. In one of his last major initiatives before retirement, Wisconsin Senator William Proxmire took the opportunity to criticize the US space program as an exercise in corporate welfare for Boeing and McDonnell-Douglas, citing a “history of corner-cutting predicated on a nutty fantasy of space industrialization,” and criticize the Reagan administration for putting control over Lifter and the S-IVC under a private corporation. Though Proxmire did not end up having much say in the investigation (having already announced his retirement and endorsed fellow Democrat Herb Kohl for the upcoming 1988 election), his criticisms set the tone for the inquiries that would follow.
As vicious as Proxmire and his associates in the Senate could get, however, their influence over NASA and STC remained limited by the most crucial difference between Magellan and Apollo 1--no one had died. As NASA’s internal accident investigation procedures took effect and a team of NASA and STC engineers and managers sat down to identify the cause of the failure, they could count on relative inattention from the public and the President taking a fairly hands-off approach.
Apollo veteran and Lifter pilot John Young, then Head of the Astronaut Office, was named Chairman of the Magellan Review Board on April 17, 1988. Like NASA’s last major accident review board, that for Apollo 13, the Magellan Review Board was staffed by astronauts, administrators, and USAF officers. For two months, the Review Board zeroed in on the cause of the accident and determined exactly why it had been allowed to happen. Within hours of the launch failure, it became apparent that the problem lay with the J-2S-2 engine on the S-IVC stage, which had apparently begun its start-up procedure, but had not completed it.
As more telemetry was analyzed, STC also returned two of the lost stage’s batch-mates to the test stand at Stennis Space Center. The lost stage, S-IVC-116, had, like most S-IVCs, never been fired after the installation of its engine. The J-2S-2 had been fired by Rocketdyne at its Santa Susana test facility, but the completed stages were not generally fired after engine installation. This cost-saving measure had been implemented early on in the program, and had been planned for second-run Saturn Vs before that program had been terminated. When subjected to full-duration testing at Stennis, neither S-IVC-117 nor -118 seemed the least bit flawed. Whatever had caused S-IVC-116 to fail, it had been unique to that engine, or to the marriage of that engine to that stage--and the engine itself had shown no anomalies when it was first fired at Santa Susana in 1986.
Following the trail of paperwork, Rocketdyne and McDonnell-Douglas engineers meticulously examined the history of every part that had gone into the engine and propellant tanks that made up S-IVC-116. Rocketdyne’s engineers finally identified the issue 43 days after Magellan’s loss, tracing it to a failure of Augmented Spark Ignition system on the engine. An electrical connection between the ASI and the engine’s control board had been improperly secured--whether through a calibration failure on the torque wrench used to fasten the bolts or a mistake on the responsible technician’s part, it had been sturdy enough to take the static test at Santa Susana, but not enough to survive the stresses of first-stage flight. The faulty connection led the ASIs to light fractions of a second later than they should have, when the combustion chamber contained more propellant than it was designed to. The result was a “hard start,” or, as such incidents were sometimes known, a “hardware-rich combustion environment.” The stage’s control systems had noticed the excessive build-up of pressure in the engine and closed the propellant feed valves, but by then the resultant small explosion had damaged both the injector plate and the ASIs beyond further operation. It was a failure that could have happened at any point in the Lifter program, but had only shown itself on this flight. The failure raised questions about what other failures could have slipped past quality control checks at Huntington Beach. Static fire testing of the stage might have revealed the failure, or more thorough testing, but the former had been deemed an unnecessary expense early in the Space Transportation System’s development program, and the latter were being slowly reduced through the years as the S-IVC continued to perform reliably and STC hunted for ways to reduce the stage’s manufacturing and test costs.
The Review Board also uncovered a somewhat lax safety culture at STC, which had been under pressure to ramp up its launch rate in 1987 and 1988 in response to the growing number of orders and the planned ramp-up of the American civil manned space program. Though the Magellan Review Board did not comment on it, accounts and memoirs published in the 1990s reveal that the organization was also attempting to proactively counter the possibility of cheaper competitors in the near future. As former USAF General James Abrahamson, former NASA Associate Administrator and then-Director of the Strategic Defense Initiative, wrote, “We pushed STC to ramp up in preparation for the maturation of SDI, but what really lit a fire under them were proposals in 1987 from the Soviets that their aerospace sector would be reorganized under Perestroika, and that they would start selling launch services themselves. That and the European progress on their Ariane successor got them going more than we could--for the first time in almost a decade, Lifter had real competitors on the horizon.” STC moved to launch more often and reduce costs further, to assure its continued dominance of the global launch market. Though the Magellan Review Board did not comment on every reason, it did conclude that quality control at STC had slipped since the start of the decade, and that a culture of arrogance had taken hold. STC’s rocket engines, after all, dated to the early 1960s or late 1950s, and had been flying for twenty years--a general sentiment had emerged that anything that could go wrong already had.
On July 22, 1988, the Magellan Review Board submitted its draft findings to NASA Administrator James Beggs, concluding that the damaged ASIs were to blame for the failed launch. The Board made a number of quality-control regulations for Rocketdyne and STC to implement. Somewhat controversially, the decision to not static-fire the completed S-IVC stages was not noted as needing to be reversed. The Review Board concluded that such testing would not have caught the failure, or indeed any other failure they had identified as particularly likely. The most likely cause of a failure that such static-testing would prevent, foreign object ingestion by the rocket turbopumps, was already effectively countered by the use of wire meshes in the fuel and oxidizer feed lines. At the end of the day, concluded the report, some failures could only be checked by flight or by painstaking inspections of every bolt on the spacecraft. While a static fire might look reassuring, it would not necessarily prove anything that previous tests did not. Instead, issues like the STS-116 ignition failure could be better caught by more rigorous quality and process control, with more extensive testing of the systems of the integrated stage short of actual firings.
The Magellan Review Board concluded that almost every quality control issue they identified could be addressed through simple procedural changes at Rocketdyne and STC--better tracking of tools, more frequent inspections--and that in any event the upcoming Dual-Engine Upper Stage (DEUS) variant of the S-IVC, with its second J-2S-2 engine, would provide sufficient redundancy that most missions could be completed even with the failure of one engine. The Board recommended that Lifter operations be suspended until DEUS stages were ready for flight, a very lax restriction on the system all things considered; DEUS was, by that point, on-track for a first flight in early 1989 anyway. As Richard Truly, a Lifter pilot who had taken a management position at STC after leaving the astronaut corps, wrote in his memoirs, “We got lucky. If we had to lose a payload, there was no better time than 1988.” Shaken, but still dedicated to their tasks, NASA and its contractors set to work preparing the Space Transportation System for its return-to-flight in spring of 1989.
The loss of Magellan, however, kicked off a small storm of controversy both within and outside of NASA, centering on whether the agency had been right to spin off Lifter operations to STC in the first place, whether the Interim Semi-Reusable System architecture had been the right choice all those years ago, and, as the dependence of Spacelab, Shuttle, and NASA’s flagship unmanned programs on Lifter was thrown into sharp relief, exactly how NASA should go forward and face the last decade of the second millennium.
Senator William Proxmire of the Senate Armed Services Committee (almost certainly unwittingly) helped lay the foundation for NASA’s new direction when he arranged a series of hearings of NASA, USAF, and STC managers in early fall 1988. The most infamous political enemy of the human spaceflight program, Proxmire had made a name for himself by criticizing government waste (particularly on scientific research he found frivolous) and excessive military spending. The temporary stand-down of the Space Transportation System presented a golden opportunity to pin a Golden Fleece Award on a program that seemed tailor-made for him.
In his capacity as a member of the Armed Services Committee, Proxmire summoned STC Chairman Harry Stonecipher to testify before the Committee. Proxmire took aim at STC’s Launch Services Contract with the Department of Defense, criticizing the company for using the hardware financed by that contract to operate a commercial launch service, and doing a bad job of that to boot.
“So, tell me. I have here this report from NASA, your biggest client. They say you took those funds, failed to pay overtime, ran your operations to the bone after every red cent, and then managed to destroy a multi-million dollar piece of NASA property because of a $25 wrench. Why should we trust that you will fix this boondoggle, that you will do anything different in the future? Why should we trust that you can economically and reliably deliver payloads for the Department of Defense?”
While this was in character for the man regarded in the spaceflight community as Senator Proxmire, Enemy of Progress, it is important to recall that his ire toward spaceflight was not all-encompassing--and indeed, that contributed to his grievances. In 1983, Promire had been persuaded by Carl Sagan to support, or at least not oppose, the Search for Extraterrestrial Intelligence. In the years since then, he had warmed to at least the unmanned part of the American space program, particularly those parts relating to Earth observation. While most writers at the time dismissed Proxmire’s attacks as cynical self-promotion, of the same kind as that which Proxmire’s hated predecessor, Joe McCarthy, employed against alleged Communists, there does appear to have been an element of genuine anger that one of NASA’s “worthwhile” missions had been lost by a failure of a vehicle in the manned spaceflight program. He spent a great deal of time asking Stonecipher whether STC had taken special precautions to ensure a successful flight for Magellan. Stonecipher, for his part, answered that every Lifter launch was taken very seriously by STC, but that every flight carried some risk. “We’re aiming for airliner-like operation. But even the 747 I took to get here from Los Angeles does not have a perfect safety record.”
Moving on from STC, Proxmire next summoned the Director of the Launch Contracts Office at NASA to explain how much oversight NASA had over STC, and why NASA had not called attention to the culture of complacency noted by the Magellan Review Board. The Launch Contracts Office, established in 1983, awarded launch contracts to STC and to Martin-Marietta (beginning in 1985 with the Complementary Expendable Launch Vehicle (CELV) block-buy). It was also the main office through which NASA interacted with STC for Lifter operations, though a separate office, the Crewed Spacecraft Launch Operations Office, coordinated Shuttle operations with STC until separation from the S-IVC, after which Johnson Space Center took over directly. Proxmire asked the Director, Timothy Cizadlo, why NASA had chosen to stick to Lifter even as this culture developed, rather than go with proven expendable launch vehicles like Titan IIIE or Atlas-Centaur, whose pads, still in mothballs, could have theoretically been revived. Cizadlo answered gracefully enough to get a laugh out of Proxmire’s colleagues: “We didn’t want to fleece the taxpayers by buying the same service for a higher price.” Proxmire, undeterred, continued by calling into question NASA’s ability to oversee even its unmanned spacecraft. “Perhaps NOAA would do a better job studying the atmospheres of other planets,” he mused.
As the hearing continued, Proxmire called into question the efficacy of the Space Lifter program in satisfying the US government’s space access needs, asking whether the program was really a great improvement over earlier, expendable rockets. He asked further whether the Launch Contracting Office was under pressure to favor STC over other programs, due to that company’s closer relations to NASA’s manned spaceflight program. On this point, Cizadlo was adamant: "One failure does not obscure the fact that the Lifter has been a success. Launch costs are down. Private investment in space is up. We've turned the investment of NASA into an entire new sector of the American economy. I challenge anyone in this room to tell me that Atlas or Titan could have done that." Hearing no objection, and seizing the moment, he went on: “My office has done business with STC since that company’s foundation, and during that time, and even before that, while Lifter was under NASA’s direct jurisdiction, the program’s safety record and costs were equal to or better than those of the expendable boosters it’s replaced. Better than Atlas or Titan could have, Lifter has enabled NASA to achieve its goals in space--and when I say that, I’m not just talking about launching any given payload, but about the objectives listed in the National Aeronautics and Space Act--the preservation of American leadership in applied space technologies. Launching cheaply is not the end goal, though it is an important part of our selection process. My job, and the job of everyone at LCO, at NASA, and the job for which we pay STC, is to expand American companies’ access to space. In my judgement, even with this recent incident, STC has done an admirable job.”
Ultimately, Proxmire’s hearings did not have a great impact on the relationship of NASA and STC, or on STC’s place as the primary launch provider for US government satellite services. While Ariane won more launch contracts in the years after Magellan than it did before, as satellite operators made sure to keep relations with Arianespace open in the event of another failure, STC would ultimately return the Space Transportation System to flight, and reclaim its share of the commercial satellite market. Contemporary political commentators wrote the hearings off as one last windmill at which Proxmire wanted to tilt before his retirement, an assessment that Proxmire himself strengthened when, in January of 1989, he awarded the last Golden Fleece of his career to the Space Transportation Corporation and NASA’s Launch Contracting Office (for his part, STC Chairman Stonecipher is said to have hung that Golden Fleece on the wall in his home office, remarking “it’s a shame they didn’t call Lifter ‘Argo’”). However, Cizadlo’s defense of the accomplishments of the Space Transportation System reflected a growing sentiment at NASA that the agency’s role in opening the High Frontier was that of a trailblazer.
Different craft. The italicized bits are narrating the first Shuttle mission, which took place a while ago relative to the rest of the post.
Note: in view of the fact that my responsive style tends to very long text and worse, I am experimenting with tucking passages that are long, involved, and/or may be regarded as peripheral under Spoiler tags, which I strive to name so as to index the general drift of what I was trying to say. These are added later and I might forget or feel pressured to skip them, or get sloppy with them. Also I feel I have to explain what I am doing since Spoiler covers are meant for another purpose. It means if I ever have a normal Spoiler I will have to call it "Real Spoiler" or something like that. None of these are Spoilers in the sense that I have any inside information on the narrative and probably any predictions or recommendations I do make will go straight to the circular file anyway. But I have had authors--not these ones yet!--ask me to back off guesses that hit too close. But in such cases if asked I would keep off the sensitive topic completely.
This is also the first time I've tried experimenting with Spoilers within Spoilers to tighten the hidden stuff up tighter, so far it seems you can bury them as many layers as you like, at least 2! It works in Preview, if not in the real post I'd have some editing to do!
The Safety Song and Dance:
Having a backup RL-10 design is, for reasons made clear to me last season, a non-starter I suppose--it would near double the design/testing cost and extend the time frame of the Lifter program. Maybe not the latter because the prime program could be tentatively approved for launch testing before the backup 16 engine alternative is done with testing, so the main design would be the time pacer, but a lot of money would have to be spent on something everyone hopes never to use.
Under the circumstances, the responsible thing for NASA and DOD to do (I assume they alone run all Orbiter missions, SLC contracting the Lifter boost for them, and that all other Lifter payloads as yet remain uncrewed except for the Lifter crew of 2) would be to pessimistically figure that future failures will happen somewhat more often than hitherto. That is, they were unusually lucky in the number of launches got in before the Magellan failure, say that the expectation of the next failure for any reason is going to be around 2/3 the number of launches they've had thus far, and revisit Orbiter crew survival features to better guarantee survival in the event of a boost failure at any point up to second stage start.
If they were to persist in sticking with the one versus two engine pair of available upper stages, they also have to figure on failure at any arbitrary point between second stage ignition and burnout. If they did something like my hypothetical 1 + 6 engine design, then if the J engine fails, the weak but available thrust of the RL engines, which burn the same propellant mix, could serve to push them on to a more useful burnout speed, one that might guarantee range enough for emergency landings in Africa--and far enough into the burn this could still be an option if some RL engines are out as well, just as they are reasoning in your post that with two J engines the situation is quite different than with just one.
Perhaps despite the cost of developing an intermediate sized mixed engine type second stage and the higher cost per kg of using it (at best, it can be comparable) it is worthwhile to develop it intending it mainly for Orbiter launches. It should raise the payload margin of the Orbiter considerably, probably beyond the Orbiter's ability to accommodate to be sure--but perhaps any further margin can be used for propellant and crew consumables to stretch the Orbiter's height and inclination range and extend mission durations.
-------------------
Overall the Lifter system is a vast improvement on the OTL Shuttle design, with the majority of emergency conditions having a plausible survival contingency for the crew versus the opposite being the case OTL-in the ATL, only the most extreme failure modes pose the low odds of survival that the best case escape options for the crew did OTL, and in cases where OTL the chances were zero the ATL system gives very good odds.
My above rant is summarized by my perception that one set of contingencies in which two crew are doomed could have been avoided, perhaps with not great survival odds but nonzero ones, for free, by a little forethought. I went over this with the authors last year and their answer is that the contingency is low probability, but I never felt it addressed my claim that even so facing it and improving it would have cost nothing in the early design phase, though now it would certainly cost a lot--basically, scrapping the old Launchers and replacing them with new ones and rearranging the tower service layout somewhat, more to the advantage than disadvantage of routine operations.
Basically to do it now would require waiting for the next generation of Lifter, and with that delayed into the 90s and possibly forever if some competing approach proves overall more cost effective, any 2nd generation Lifter will probably omit the human flight crew completely, in favor of more advanced avionics allowing the whole flight profile to be managed by a combination of advanced autopilot and coaching from remote ground control. The most critical flight operation is landing the thing which will always happen with the Lifter close to the landing field and getting closer by the second, so securing adequate closed loop communications with negligible time lag will not be a problem, and any inadequacies can be addressed largely by doubling up on ground equipment numbers and power.
So if the odds of this mode of failure are 1/5000 or so then it is hardly likely to show up. Later failures of say 1/200 odds should dominate the narrative I suppose, and every crew member in every launch has excellent chances of surviving the later failures. I feel my suggestion of a 1.3 smaller upper stage is admittedly costly but does improve those odds to something better still for the Orbiter crew, at a price to be sure, but also with some nearly compensating payoffs in superior performance envelope for a nominally successful mission.
This is also the first time I've tried experimenting with Spoilers within Spoilers to tighten the hidden stuff up tighter, so far it seems you can bury them as many layers as you like, at least 2! It works in Preview, if not in the real post I'd have some editing to do!
The Safety Song and Dance:
I looked at the RL-10A-3A rocket engine, which if I can trust Encyclopedia Astronautica dated from 1984 and might therefore have something comparable on the shelf at the time of the 1998 Magellan launch failure. To match the thrust of a single J-2S (assuming that is reported correctly and that the ATL late-80s version still matches the vacuum thrust spec of 1138.5 kN) sixteen of the small but efficient and reliable engines would be required. The higher Isp of 444 sec, due mainly to the extended nozzle, would perhaps allow a few of those to be left off, or compensate enough for several engines out I suppose. But the thrust has to be in the right ballpark initially for the upper stage to do its job.
I think then it is a possible contingency if the J-2S engine is suspected might be to have an alternate engine set version designed and ready for testing, but with so many RL-10s being disposed of in one shot the economics would look dubious even aside from the fact that finalizing the construction design, making stages for ground testing. I am not sure if ground firing RL-10 designs mounted to stages even works at sea level, let alone whether the data are relevant to a high altitude burn--to be sure the thrust and gas flow are of the same magnitudes as a single J-2S and therefore if a chamber or extrapolation procedures valid for the J engines exists, a 16 RL-10 ensemble should be possible to test as well as a single J engine.
Of course sixteen engines takes us well past Saturn 1 or Falcon 9 territory well on the way to the oft-denounced N-1's minimum complement of 24 engines, OTL not even attempted until they had been expanded to 30. It is 2/3 of the way in numbers to the original sketch of 24 and over halfway to the eventual count of 30. I do believe N-1 could have been made to work, with an earlier start and more modest goal, sticking to the original plan of 24, in part because the extra six appear to have been the cause of much of the trouble and withal would only have been used for 30 seconds anyway, but of course the Soviet Ker-Lox engines, which had quite high Isp and ambitious closed cycle staged combustion design, were nowhere near as developed in terms of reliability as the workhorse Centaur engine had come to be OTL by 1988, and I believe it was used comparably as much ITTL, probably even on Lifter payloads. Still, sixteen engines represents an invitation to have one or two fail and perhaps in a fashion causing a cascade of failures.
At 2256 kg for 16 of them, the RL-10 alternative would also mass 61 percent more than a single J-2S engine (dry weights being compared in each case) which, barring advantages from the higher Isp (versus 436 for the J engine) subtracts 856 kg from the payload too. Also if I figure right, an ensemble of 16 RL-10s would draw more propellant per second at full thrust than the J engine would--a very small percentage to be sure, 3.446 kg more versus 266.2 per second for the J engine. In fact the exact thrust ratio is 15.51 not 16, so starting with 16 gives a little extra thrust initially--more than enough to compensate for greater engine mass. But engines have to be mounted and integrated into thrust structures, and it is not clear how much more mass must be allowed for the thrust structure. With so many engines, we might get same savings--gimballing only the ones on the outer rim, and only in one dimension each for tangent vectoring might be overall lighter than having the gimbal the single J-2S in two dimensions--also, there would need to be some sort of vernier engine to control roll, I suppose this might have involved installing a couple RL-10 anyway along with the J engine. Anyhow, with a need for 16 and a bit of extra thrust, some 3 tonnes or so extra, to start with, shutting down some engines later in the burn deliberately would bring the later rate of consumption down, just a few could more than compensate and leave the RL-10 version perhaps more efficient overall.
I think then it is a possible contingency if the J-2S engine is suspected might be to have an alternate engine set version designed and ready for testing, but with so many RL-10s being disposed of in one shot the economics would look dubious even aside from the fact that finalizing the construction design, making stages for ground testing. I am not sure if ground firing RL-10 designs mounted to stages even works at sea level, let alone whether the data are relevant to a high altitude burn--to be sure the thrust and gas flow are of the same magnitudes as a single J-2S and therefore if a chamber or extrapolation procedures valid for the J engines exists, a 16 RL-10 ensemble should be possible to test as well as a single J engine.
I gather from all the boasting of superior NASA extensive engine test procedure versus Soviet shortcut practices, NASA must have some way to do valid vacuum equivalent testing on upper stage engines, all I can imagine is a vast vacuum chamber with massive cooling facilities and a pumping system that can remove hundreds of kilograms of very hot exhaust products as fast as they are dumped into the chamber, to maintain the near-space vacuum conditions--after all even second stages are going to be lit when the air is not quite hard vacuum yet so a certain amount of test chamber gases persisting is acceptable--one could spray cold water to reduce the temperature many tens of meters downstream though that would compound the pumping requirements and make the thin "air" in the chamber mostly water vapor--at low high stratospheric pressures water will be a gas at reasonably low temperatures to be sure. But a gas with different dynamic reaction properties than thin air, so that has to be either accounted for or of verifiably negligible importance.
Having a backup RL-10 design is, for reasons made clear to me last season, a non-starter I suppose--it would near double the design/testing cost and extend the time frame of the Lifter program. Maybe not the latter because the prime program could be tentatively approved for launch testing before the backup 16 engine alternative is done with testing, so the main design would be the time pacer, but a lot of money would have to be spent on something everyone hopes never to use.
However I wonder about this--an intermediate thrust design made with 1 J-2S and say 6 RL-10 could have some interesting advantages. With 6 small engines mounted with very high angles of gimbal in one dimension, I think sufficient yaw-pitch control authority could be provided by the 6 "verniers" alone, in fact pairs of engines out might still leave it degraded but adequate on 4 engines. Then the J engine could have a very simple fixed mount which should save cost, weight and make it more reliable.
In addition, using 2 or 3 (or if desired, all 6) RL engines as ullage engines can enable simplifications in the J engine start/restart system. The RL-10 family began to incorporate repeated restart capabilities early on and they are standard in the series so in this role they should be very reliable. Of course the J-2 was designed out of the box for at least one restart so failure to start the first time is a pretty surprising failure mode. Given that this is where it went wrong on the Magellan launch, I wonder if anyone at NASA or STC is looking at a hybrid RL-10/J-2 based system now.
Part of the whole genius of the Lifter system is to save money on the expendable second stages and multiplying the engine installation by 7 (or a lesser fraction--the second stage did require some sort of vernier engine for roll control in the standard design, does it not? Gimbaling RL-10s would fill that role with capacity to spare) is going to be costly--but I think only to the tune of the relative masses, and perhaps less than that for RL-10 engines.
Assuming J engines and associated structural masses cost 2/3 the dollars J engines do per kilogram, and equating the ratio of required thrust structure mass and cost (I proposed a very elaborate high angle and tightly controlled gimbal, but in only one dimension, which should save some thrust structure mass and eliminate gimbaling of the J engine, or perhaps anyway save on both with a more limited degree of freedom) the added 6 engines would add 50 percent to the cost of the engine section. With the optimistic assumption of 1/3 per kg cost, especially unlikely factoring thrust/gimballing structures in, that comes down to 25 percent more, so the engine portion of a second stage would cost between 9 to more likely 30 percent more per tonne placed in orbit. That in turn is not going to be the whole cost of a launch but it will raise overall costs by some 10 percent. We'd have to raise payload by a more aggressive 30 percent to break even versus the old system, which would fall short if anything fails.
In addition, using 2 or 3 (or if desired, all 6) RL engines as ullage engines can enable simplifications in the J engine start/restart system. The RL-10 family began to incorporate repeated restart capabilities early on and they are standard in the series so in this role they should be very reliable. Of course the J-2 was designed out of the box for at least one restart so failure to start the first time is a pretty surprising failure mode. Given that this is where it went wrong on the Magellan launch, I wonder if anyone at NASA or STC is looking at a hybrid RL-10/J-2 based system now.
Part of the whole genius of the Lifter system is to save money on the expendable second stages and multiplying the engine installation by 7 (or a lesser fraction--the second stage did require some sort of vernier engine for roll control in the standard design, does it not? Gimbaling RL-10s would fill that role with capacity to spare) is going to be costly--but I think only to the tune of the relative masses, and perhaps less than that for RL-10 engines.
The small engines use a much lower chamber pressure (again if I can trust EA, just over 32 atmospheres versus more like 75-80 for the J engines) and a very elegant, simple expansion turbo pump drive, so the labor costs should be lower per kilogram and the fact they are used in quantity annually even before a hypothetical Lifter adoption of them should make the per kilogram purchase price lower still. Six would add about 850 kg, plus say another 850 for 1700 kg for the gimbaling and thrust structure, but also add nearly 45 tonnes or almost 37 percent to gross thrust, allowing the standard tankage/payload thrust structure/payload mass to grow by a comparable amount. Let's say we limit payload growth to just 15 percent, allowing for the RL-10s to suffer occasional pairs of engines out (if the J engine fails we are screwed anyway, unless it does so very late in the burn--a benefit but a rarely needed one, we trust) and bearing in mind that adding tonnage to the upper stack lowers the burnout speed of the Lifter stage--less than we might guess than for an expendable booster since any slowdown means less need for propellant ballasting the Lifter stage, so is offset by appropriating ballast propellant to the prime boost burn. Assuming existing high volume production (by astronautical biz standards, a dozen a year or so versus one or two every several years for less ubiquitous engines) already gives a discount on top of its basically simpler, less demanding design, and that STC ordering at least six for every projected Lifter launch in the future allows a further volume discount, I would guess it would be conservative to guess they cost less than a J engine kilogram for kilogram by a factor of 2/3. Really it might be more like 1/3 with the volume orders STC would be putting in but let's stick with the less favorable figure.
Assuming J engines and associated structural masses cost 2/3 the dollars J engines do per kilogram, and equating the ratio of required thrust structure mass and cost (I proposed a very elaborate high angle and tightly controlled gimbal, but in only one dimension, which should save some thrust structure mass and eliminate gimbaling of the J engine, or perhaps anyway save on both with a more limited degree of freedom) the added 6 engines would add 50 percent to the cost of the engine section. With the optimistic assumption of 1/3 per kg cost, especially unlikely factoring thrust/gimballing structures in, that comes down to 25 percent more, so the engine portion of a second stage would cost between 9 to more likely 30 percent more per tonne placed in orbit. That in turn is not going to be the whole cost of a launch but it will raise overall costs by some 10 percent. We'd have to raise payload by a more aggressive 30 percent to break even versus the old system, which would fall short if anything fails.
Under the circumstances, the responsible thing for NASA and DOD to do (I assume they alone run all Orbiter missions, SLC contracting the Lifter boost for them, and that all other Lifter payloads as yet remain uncrewed except for the Lifter crew of 2) would be to pessimistically figure that future failures will happen somewhat more often than hitherto. That is, they were unusually lucky in the number of launches got in before the Magellan failure, say that the expectation of the next failure for any reason is going to be around 2/3 the number of launches they've had thus far, and revisit Orbiter crew survival features to better guarantee survival in the event of a boost failure at any point up to second stage start.
If they were to persist in sticking with the one versus two engine pair of available upper stages, they also have to figure on failure at any arbitrary point between second stage ignition and burnout. If they did something like my hypothetical 1 + 6 engine design, then if the J engine fails, the weak but available thrust of the RL engines, which burn the same propellant mix, could serve to push them on to a more useful burnout speed, one that might guarantee range enough for emergency landings in Africa--and far enough into the burn this could still be an option if some RL engines are out as well, just as they are reasoning in your post that with two J engines the situation is quite different than with just one.
Perhaps despite the cost of developing an intermediate sized mixed engine type second stage and the higher cost per kg of using it (at best, it can be comparable) it is worthwhile to develop it intending it mainly for Orbiter launches. It should raise the payload margin of the Orbiter considerably, probably beyond the Orbiter's ability to accommodate to be sure--but perhaps any further margin can be used for propellant and crew consumables to stretch the Orbiter's height and inclination range and extend mission durations.
-------------------
I am much relieved the first major glitch in the Lifter program was not in the Lifter stage itself. The crew have that nifty escape capsule, but I already registered my fears that sudden unexpected catastrophe on the launch pad itself is a fairly high probability portion of the set of possible Lifter failure contingencies, and that with the current layout of the flight deck on the dorsal side facing the service tower instead of the opposite direction, any sudden emergency then will doom the two Lifter crew since they would not have time to bail out on jump lines a la OTL's pathetic escape mode, nor does the capsule do them any good since it will slam them straight into the service tower.
Whereas I would think that a capsule that could get them far enough from a nearly full Lifter shortly after clearing the tower would be able to angle them high enough as well as far enough laterally that there would be time for their ejection seats to save them--I am not sure if ejection from the capsule would always be required in which case the capsule as a whole would have parachutes and be designed to hit ground or water survivably with crew aboard--that would be better since they need to be protected from the initial blast and ideally staying inside a floating and generally heavily survival equipped capsule is superior to parachuting with an inflatable raft into the ocean or some Floridian's back yard.
I think the capsule is bloody brilliant versus OTL, even if the pair must abandon it before they hit the surface; standard issue ejection seats from there are good enough I guess though designing to land in the capsule would be better. But as I've said before the maddening thing about the gap in its use in case of sudden catastrophic breakdowns before it has cleared the launch service tower is that this contingency too would be very well covered by simply turning the Lifter's pad orientation around.
At first glance this would seem tonrequire either awkward reach around to side or ideally dorsal loading ports. But consider: the main crew hatch is already on the side so a tangential extension would be--no change at all really!
Extending the main propellant lines is perhaps more of not just hassle and cost but also added risk, in mind of last year's SpaceX pad loss which turned out to be propellant line related.
But, such dense flows can very reasonably have been designed to go through ventral hatches in the TPS belly--landing gear hatches are needed there after all, and with the tail on the dorsal side (this post began by reminding us about them after all) the Lifter could have been more snugly placed to the tower ventral than dorsal anyway, actually shortening the feed lines if they went through belly hatches. (Or with lines the same extension and thus risk as chosen in the ATL, the lower stage clearance is superior to the dorsal to tower arrangement). The OTL Orbiter required ventral feed of its hydrogen and oxygen from the ET after all and those required hatches--perhaps in some ATL somewhere, an infamous loss of Orbiter case came about through such hatches failing to close or having the block of tiles over them wrecked due to being dinged or ripped during separation from the tank, but OTL they worked fine. As did the landing gear hatches which the Lifter still requires, and for gear landing a considerably greater weight too.
With the Lifter suitably turned around--Right Side, one might say--I'd feel great about the Lifter crew's odds of surviving even the most far fetched failure scenario. As it is, it looks to me like a risk foolishly taken just for the sheer hell of it, when the engineering alternative that would cover it costs absolutely nothing more than a little foresight and pessimism.
So--I have been waiting for the fatal Loss of Crew event in which an Orbiter mission atop a Lifter successfully clears the tower when the Lifter suddenly blows up with no more than 3/4 of a second warning (just as SpaceX's launcher did last year, recall) using its over engineered emergency escape engines, but the Lifter crew dies in the blast, the capsule eject system disabled due to it being aimed at the tower and the two man Lifter crew nevertheless cannoned into it and crushed when the capsule flattens like an empty Sprite can in the blast--oh, and despite the capsule eject engines not lighting and stoically sitting unignited throughout the pad fire, the general blast brings down the tower anyway. The Lifter holds a lot more propellant than that Falcon 9 did after all! If at some point the capsule pyrotechnics do cook off as one would reasonably expect them to, that will add a pair of nasty Roman Candles to the general mess. Yay on saving six crew, boo on losing two. If it just faced the other way, even if the Lifter crew are killed before their capsule can get far enough away, at least the extra damage the capsule pyrotechnics would do at the site is removed to a separate location.
Far more likely any Lifter failure will happen later, of course. But I do believe those initial seconds of startup, like the critical moments of takeoff and landing with an airplane, are inherently more fraught with risk per second than during the middle of a burn, also that very sudden unexpected failures that veer out of parameters faster than abort methods can shut them down are considerably more likely during startup than during a sustained burn. It may be that with the crew capsule pointed the right way there might not be time to activate the eject sequence before the blast catches up with it, but at least it will be batted in the right direction if separation is barely accomplished.
Therefore it may be that odds are lower of failure then than during boost overall, but not in proportion to the relative time spans involved. It may be that the risk of failure before the capsule clears the tower is only 1 chance in 20 out of 19 more probable paths to failure, and since the overall odds are figured at something like 1 in 500 or so, it is not anticipated that such an extreme case will happen.
But again I say--it would have cost essentially nothing to simply design the Lifter to face the other way on the pad, and improve those survival chances by an admittedly small increment--for free.
Whereas I would think that a capsule that could get them far enough from a nearly full Lifter shortly after clearing the tower would be able to angle them high enough as well as far enough laterally that there would be time for their ejection seats to save them--I am not sure if ejection from the capsule would always be required in which case the capsule as a whole would have parachutes and be designed to hit ground or water survivably with crew aboard--that would be better since they need to be protected from the initial blast and ideally staying inside a floating and generally heavily survival equipped capsule is superior to parachuting with an inflatable raft into the ocean or some Floridian's back yard.
I think the capsule is bloody brilliant versus OTL, even if the pair must abandon it before they hit the surface; standard issue ejection seats from there are good enough I guess though designing to land in the capsule would be better. But as I've said before the maddening thing about the gap in its use in case of sudden catastrophic breakdowns before it has cleared the launch service tower is that this contingency too would be very well covered by simply turning the Lifter's pad orientation around.
At first glance this would seem tonrequire either awkward reach around to side or ideally dorsal loading ports. But consider: the main crew hatch is already on the side so a tangential extension would be--no change at all really!
Extending the main propellant lines is perhaps more of not just hassle and cost but also added risk, in mind of last year's SpaceX pad loss which turned out to be propellant line related.
But, such dense flows can very reasonably have been designed to go through ventral hatches in the TPS belly--landing gear hatches are needed there after all, and with the tail on the dorsal side (this post began by reminding us about them after all) the Lifter could have been more snugly placed to the tower ventral than dorsal anyway, actually shortening the feed lines if they went through belly hatches. (Or with lines the same extension and thus risk as chosen in the ATL, the lower stage clearance is superior to the dorsal to tower arrangement). The OTL Orbiter required ventral feed of its hydrogen and oxygen from the ET after all and those required hatches--perhaps in some ATL somewhere, an infamous loss of Orbiter case came about through such hatches failing to close or having the block of tiles over them wrecked due to being dinged or ripped during separation from the tank, but OTL they worked fine. As did the landing gear hatches which the Lifter still requires, and for gear landing a considerably greater weight too.
With the Lifter suitably turned around--Right Side, one might say--I'd feel great about the Lifter crew's odds of surviving even the most far fetched failure scenario. As it is, it looks to me like a risk foolishly taken just for the sheer hell of it, when the engineering alternative that would cover it costs absolutely nothing more than a little foresight and pessimism.
So--I have been waiting for the fatal Loss of Crew event in which an Orbiter mission atop a Lifter successfully clears the tower when the Lifter suddenly blows up with no more than 3/4 of a second warning (just as SpaceX's launcher did last year, recall) using its over engineered emergency escape engines, but the Lifter crew dies in the blast, the capsule eject system disabled due to it being aimed at the tower and the two man Lifter crew nevertheless cannoned into it and crushed when the capsule flattens like an empty Sprite can in the blast--oh, and despite the capsule eject engines not lighting and stoically sitting unignited throughout the pad fire, the general blast brings down the tower anyway. The Lifter holds a lot more propellant than that Falcon 9 did after all! If at some point the capsule pyrotechnics do cook off as one would reasonably expect them to, that will add a pair of nasty Roman Candles to the general mess. Yay on saving six crew, boo on losing two. If it just faced the other way, even if the Lifter crew are killed before their capsule can get far enough away, at least the extra damage the capsule pyrotechnics would do at the site is removed to a separate location.
Far more likely any Lifter failure will happen later, of course. But I do believe those initial seconds of startup, like the critical moments of takeoff and landing with an airplane, are inherently more fraught with risk per second than during the middle of a burn, also that very sudden unexpected failures that veer out of parameters faster than abort methods can shut them down are considerably more likely during startup than during a sustained burn. It may be that with the crew capsule pointed the right way there might not be time to activate the eject sequence before the blast catches up with it, but at least it will be batted in the right direction if separation is barely accomplished.
Therefore it may be that odds are lower of failure then than during boost overall, but not in proportion to the relative time spans involved. It may be that the risk of failure before the capsule clears the tower is only 1 chance in 20 out of 19 more probable paths to failure, and since the overall odds are figured at something like 1 in 500 or so, it is not anticipated that such an extreme case will happen.
But again I say--it would have cost essentially nothing to simply design the Lifter to face the other way on the pad, and improve those survival chances by an admittedly small increment--for free.
Overall the Lifter system is a vast improvement on the OTL Shuttle design, with the majority of emergency conditions having a plausible survival contingency for the crew versus the opposite being the case OTL-in the ATL, only the most extreme failure modes pose the low odds of survival that the best case escape options for the crew did OTL, and in cases where OTL the chances were zero the ATL system gives very good odds.
My above rant is summarized by my perception that one set of contingencies in which two crew are doomed could have been avoided, perhaps with not great survival odds but nonzero ones, for free, by a little forethought. I went over this with the authors last year and their answer is that the contingency is low probability, but I never felt it addressed my claim that even so facing it and improving it would have cost nothing in the early design phase, though now it would certainly cost a lot--basically, scrapping the old Launchers and replacing them with new ones and rearranging the tower service layout somewhat, more to the advantage than disadvantage of routine operations.
Basically to do it now would require waiting for the next generation of Lifter, and with that delayed into the 90s and possibly forever if some competing approach proves overall more cost effective, any 2nd generation Lifter will probably omit the human flight crew completely, in favor of more advanced avionics allowing the whole flight profile to be managed by a combination of advanced autopilot and coaching from remote ground control. The most critical flight operation is landing the thing which will always happen with the Lifter close to the landing field and getting closer by the second, so securing adequate closed loop communications with negligible time lag will not be a problem, and any inadequacies can be addressed largely by doubling up on ground equipment numbers and power.
So if the odds of this mode of failure are 1/5000 or so then it is hardly likely to show up. Later failures of say 1/200 odds should dominate the narrative I suppose, and every crew member in every launch has excellent chances of surviving the later failures. I feel my suggestion of a 1.3 smaller upper stage is admittedly costly but does improve those odds to something better still for the Orbiter crew, at a price to be sure, but also with some nearly compensating payoffs in superior performance envelope for a nominally successful mission.
TheHolyInquisition
Banned
Well, come November we should have more data on large-engine-count rockets.
Of course, this discounts the mentioned RL-10 stage being a balloon system.
Well, the only failure thus far has been on the edge of atmosphere at second stage ignition.
Hmm. If they really wanted a challenge in advancement, they could try to build an F-2 engine to Merlin levels of TWR. That should be more than enough challenge for the "improved engines" lobby, and a doubled TWR from current is someone engineers would enjoy. Even without that unlikely move, though, the next-gen lifters will be very impressive.
@Shevek23 , a few points in reply:
If you're curious, here's a paper on the altitude testing of the J-2S(-1) historically, and on the J-4 Test Cell which carried out the original altitude testing for the J-2S, as well as for other engines. For typical acceptance firings, the simpler test arrangement is to remove the vacuum bell extension, allowing firing the engine at sea level, then comparing to expected performance figures for that configuration.
The J-2S-2 as used on Lifter's SIV-C is not being called on for a tightly configured thrust structure like the S-II, and is intended to be used for GTO missions.Thus, the original J-2S-1 is re-engineered into a J-2S-2 with a larger expansion ratio. I don't have our exact expansion numbers in front of me (it's I think either 60 or 80), but the result is an ISp of about 450, equivalent or better than the comparable vintage RL-10. Thus, the two-engine S-IVD "DEUS" stage delivers better efficiency than the original S-IVC, largely thanks to higher initial T/W than either the S-IVC or your proposed 6-and-1 configuration, and more predictable engine-out performance.
If you're curious, here's a paper on the altitude testing of the J-2S(-1) historically, and on the J-4 Test Cell which carried out the original altitude testing for the J-2S, as well as for other engines. For typical acceptance firings, the simpler test arrangement is to remove the vacuum bell extension, allowing firing the engine at sea level, then comparing to expected performance figures for that configuration.
The J-2S-2 as used on Lifter's SIV-C is not being called on for a tightly configured thrust structure like the S-II, and is intended to be used for GTO missions.Thus, the original J-2S-1 is re-engineered into a J-2S-2 with a larger expansion ratio. I don't have our exact expansion numbers in front of me (it's I think either 60 or 80), but the result is an ISp of about 450, equivalent or better than the comparable vintage RL-10. Thus, the two-engine S-IVD "DEUS" stage delivers better efficiency than the original S-IVC, largely thanks to higher initial T/W than either the S-IVC or your proposed 6-and-1 configuration, and more predictable engine-out performance.
Last edited:
But sadly, as here, not quite as reliable engine-in performance as they hoped!
Note that I gave you post a like, so I am just trying to clarify--I was proposing not a replacement for the DEUS two engine version, but an upgrade/alternate standard "single" engine lighter configuration--since it turned out that relying 100 percent on the single J-2S-2 led to total mission failure. I considered but rejected the notion of replacing the single engine with sufficient numbers of RL-10 to match it. To be totally honest, I did the math in part because so often in the past half year or so of the hiatus, we've crossed swords where you'll pull out the RL-10 as a wonderful alternative to say the SSME. Since the SSME is even higher thrust than the J series, I trust that the reducto ad absurdum for bringing in the RL-10 as though it were a relevant alternative in cases where the mission requires engines in the thrust ballpark of a J engine or SSME type will be remembered? I concluded that simply swapping in RL-10s versus using engines like J engines is like proposing to replace a plow horse with suitably trained mice. Or anyway chihuahuas. Mush!
Note that if I had checked the back story and reminded myself the J-2S-2 is upgraded in ISP with an even bigger expansion ratio to SSME/later generation RL-10 levels of 450 sec that that versus 436 sec for the OTL version implies higher thrust--I just did the math, apparently 16 RL-10s would fall short instead of exceeding the requirement by half an RL-10 thrust--16.1 RL-10 versus 15.5 would be needed. So actually it doesn't change anything I wrote much except that that particular version of RL-10 is not superior, but slightly inferior, to the -2 J engine you describe here, in Isp.
Having worked that out with a subtext and an agenda, I did then consider how judicious use of modest numbers of RL-10s might help raise confidence in the single J engine version of the S-IV family of upper stages. Six might also be excessive too; if one trusts the engines will not fail as few as three flanking the singe J engine will be enough to exploit most synergies I suggested--though with just three little RL-10s I guess there is no way their gimbaling could relieve the central J engine of its own need to gimbal, and in two dimensions--but very modest one dimensional gimbaling of the three RL-10s could certainly give very brisk roll control while hardly lowering the useful forward thrust at all. That means that between their low thrust and modest gimbaling requirement the multiplier of their dry mass in figuring the mass penalty versus slightly raised thrust can be very low.
I guess better ullage settling prior to attempting to ignite the J engine will not in fact have any bearing on making that ignition more reliable; either they QC the upper stage components right or they don't, and of course expecting perfect reliability every time is Utopian and attempting it can be so costly in processing time hence labor/lab costs that it eats up the whole margin of Lifter launches and more, and their accomplished track record is quite as good as any EELV developed thus far--whereas competitors attempting economies via recovery and reuse will introduce new costs as well as new opportunities for glitches that render their predicted reliability very speculative indeed. Anyway if ullage settling or simplifying the J engine mount by putting attitude control on the RL engines is irrelevant to the problem at hand, then the added reliability I hoped to offer by adding the RL engines is nonexistent and of course adding engines adds more possible points of failure. I did think maybe adding the RL engines might address reliability concerns for the crewed Orbiter launches, but in fact this can probably be better addressed either by, as I did stress, relying on the escape options the ATL Orbiter has that are already built in, or conceivably by switching future Orbiter operations over to the DEUS stages, which have two separate J engines already. Of course that means wasting the higher price and heavier mass of the bigger upper stage, unless one compensates by adding payload to go with the Orbiter piggyback, in either an independent mission payload bus behind the Orbiter--or integrating some or all of that with the Orbiter mission by making it a sort of "trailer" appendix to the Orbiter, a mission module to either be delivered to some station or free flying to accomplish a mission comparable to OTL Space Lab. Such a mission module would presumably be expendable, though the option of attempting to make it recoverable with some sort of independent Earth return strategy would also exist. But one could always simply launch an Orbiter on a DEUS stage with scanted propellant. It costs more but if one engine fails and the other does not, it might have at least the needed boost to reach an African emergency landing field instead of having to either attempt a loop abort on OMS propellant back to the launch site or manage to bail the crew out before the Orbiter hits the ocean and is lost.
Which is why I went on about the Lifter crews' abort options too.
This week's installment, so cleverly linked to the "Retroburn" concept in the title, is about failure and recovery from it after all.
Now then I trust soon we'll have some retrospective summation of the success of the Lifter program as the promised "cheaper path to orbit" that it, and OTL STS, were both primarily supposed to be. IIRC the goal has been to halve the cost to orbit versus EELVs and I believe it has been said this has been accomplished in the ATL.
Now I think that if this is so, even with reliability working out to be about the same as the better EELVs but no better than that, that the global launch market, at least its entire American contingent and the foreign private sector as well, would flock to STC to use Lifter for their payloads. I think, even in the context of the sobering and costly (mostly in terms of opportunity costs of being unable to meet commitments while the system is stood down for review) Magellan failure, halved launch costs are just plain compelling. It is now seven years since the program was experimental and being tested, and it is now as standard a method of launching as any competitor--for surely the Deltas and Titans and Atlases of 1988 are not the same rockets as of 1980, and the foreign competition is also in the position of having to field new editions with no long track record in the current version.
Even knowing now that STC cannot guarantee perfect success every time, that is no less true of the alternatives and any new competitor claiming better in that respect is speculating at best, if not prevaricating. Setting the safety issue as moot because it is equivalent across the board (and I gather so many Lifter missions have flown that even counting the Magellan loss, Lifter is objectively superior to most in track record) then the question of total launch cost is pretty decisive.
It is clear that overall, despite not inconsiderable refurbishment costs and that Lifter is a much heavier system on the pad, it costs STC half as much as any EELV would have to charge to break even to put a given payload up. Now the question is, how much do typical industry payloads of a given size to a given orbit cost in themselves, relative to a conceptual "standard" EELV launch cost? If the price any customer must pay in total for a revenue earning or given level of utility non-profit payload of a given size to a given orbit is as large for the hardware and its maintenance in service as the launch costs would be without Lifter, then using Lifter instead results in only a modest total savings--25 percent versus the cost of launching it the old fashioned way and then operating it in orbit. But if these strictly payload costs are generally small compared to the traditional launch costs, then halving those is a dramatic reduction--not quite down to 50 percent, but close. I feel that if this were the case, then every customer who could avail themselves of STC would do so. At the very least, it saves the company contracting the launch a big chunk of investment money they could deploy elsewhere. But for revenue earning satellite schemes in particular, being able to get the thing up there for little more than half the cost of the old way means they have very nearly the funds to launch a second spacecraft lying in their accounts. A firm that just pockets that savings will still be earning the same revenue as if they had launched on an EELV but having done it at half the cost, or little more, they nearly double the rate of profit on the investment. This suggests to me that firms interested in profit (and aren't they all supposed to be?) will think very hard about using the leftover savings, plus a small seed more if payload costs have traditionally been a small fraction of the whole, to launch a second spacecraft and double their revenue in this sector, still at the same very high rate of return on investment. With doubled revenues, they can more rapidly accrue the funds for a third launch much earlier than returns on just the one pre-Lifter economics restricted them to. If they don't do it, it seems likely some competitor will pretty soon, whereas if the competitor is going to come in anyway, they must attempt to secure their lead by launching in volume to stay ahead of the game. The competitor of course would be crazy to use any other system but Lifter. Barring considerations of national security and pride which could divert some competitors to using rival overseas systems, which are all at this point still essentially EELVs too, with the sole developed exception being operated by the Soviet Union. I might dream of competent Bolshevik engineers with good quality control emerging someday, being the socialist I am, but who in the competitive corporate Western world will believe that day has dawned in Russia yet? The Soviets can of course lowball the hell out of their asking price, being a state funded monopoly, but using the Soviet alternative, how soon will it be before a loss as costly as Magellan, or something much worse, happens? They already crashed their flyback booster stages the first time they tried their new system, what else is lurking in the works waiting to go wrong?
So--it seems far and away likely to me that before Magellan, that STC was finding demand for Lifter launches was already well above capacity. I gather STC was not spun off completely from government regulation, even control, and they are not free to jack up their prices to whatever the market will bear--which would be, at the upper limit, only slightly lower than EELV costs, which would mean STC could pocket enormous super profits. I gather that no, when Congress granted them the right to use Lifter the corporation had to observe stipulated limits in their profit rates, and therefore they cannot raise the price of a launch much higher than the cost to the Company to fund doing one. I think a ten percent markup at most would be the upper limit--unless aerospace companies with a successful weapons system being procured in mass routinely enjoy much higher ones. Whatever prevails in those circles over the longer terms would probably set the specified rate. If so, then I suppose the cited half price of an EELV launch is including STC's permitted profit level, just as Martin and General Dynamics would surely include their profits in their asking prices for Titan or Atlas-Centaur launches. STC is effectively mandated by a corporate charter hammered out in negotiations with the Federal Government that contracted for Lifter to be developed to pass on the savings the semi reusable platform enables to the customer, and the customer who does not take advantage of this amazing deal must surely suffer for it. I would think that if STC cannot simply ration the volume of launches by jacking up prices and pocketing super profits, they must either be operating at very high volumes of launches, or are forced to ration them by means other than price, by waiting lists and simple refusal to consider deals not on their top priority list--DoD I imagine takes top priority, followed by NASA missions when that agency has the cash in hand, then preferred Defense contractors followed by other US corporations with clean noses as far as national security goes. Perhaps agencies of favored foreign governments might take precedence less connected US private applicants.
If they were to expand to accommodate all who approached them, STC would wind up handling all launch but a loyal and well compensation portion of foreign private and the majority or entirety of foreign governmental launches by nations that have their own launcher program, and furthermore the volume of those who launched with STC would have grown considerably versus OTL in their category; by a factor of two would be surprisingly low I would think--for that would mean that the amount of funding devoted to private sector and publicly funded space travel has not increased in the slightest despite the lower price and the herd mentality of businessmen who can observe their competitors stampeding to the sky. Since revenues from profitable space based enterprises must increase with added hardware placed in orbit at half price, this must mean the investors merely milk their prior fixed investment at a doubled rate of profit but do not seek to double down and seek more where that comes from, which strikes me as ASB behavior indeed for businessmen! I think even if it is exotic and expensive to operate there and quite risky too, the higher revenue becomes an obvious sure thing; the firms that would be drawn in first and rewarded with higher profits first would be those already familiar with these costs and risks and from their point of view both have come down dramatically. If the price of a launch is halved, I think the total volume of launch customers who would pay that price or more, never need further incentives to seek opportunity at that price, would be something more like a factor of three or more the number of launches reasonably projected to have gone up on EELV--total investment should grow by at least 50 percent in the field if not more, and all of that revenue in the form of tickets to orbit purchased by this expanded sector would all be reaped by STC--provided of course they can meet the launch rate required!
If Magellan's failure clouds the issue, it will not be because investors are scared off by a possible loss risk they had never considered. Post-Magellan if not before I would recommend STC short-circuit the whole thing with a promise to pony up full value of a lost payload plus a free launch as soon as an available Orbiter matches someone's required launch window. But even if they don't do that, launch insurance is a thing OTL and was in the 1970s and 80s already and the prudent customer will seek coverage of the downside one way or another. No, it would be fear that perhaps the government will never recertify Lifter for launching again and the gravy train is derailed forever.
I would think all these firms who were able to get launch dates with STC would form one vast and powerful lobby demanding the ATL STS be put back in service ASAP and if more public funds are needed to enable that, or handle other anticipated risks proactively, they will twist arms in Congress to get the necessary development done.
The biggest danger to STC and the Lifter system in the longer run is that someone will come up with something even more economical to scoop STC in turn, but that cannot happen without some major investment and a lot of development time. The next biggest danger that can kill them, or set them up for failure by competition from a rival system with few advantages or even one somewhat inferior--but available--is if STC does not plan for the surge in volume and one way or another do what it takes to enable Lifter launches at the rate the market will desire. If they fall a little bit short that is a foot in the door for a close or superior competitor--if they fall grossly short the perception may arise that they are really only able to offer low prices due to government subsidy and the pro-Lifter lobby may reverse itself and join in with the likes of Proxmire in denouncing and shutting it down, on the pretense of safety issues if necessary.
Although I'm not writing this timeline I have been serving as a prereader, and without giving too much away, Shevek, I think that many of your questions will be addressed in upcoming posts, though it may be some time until the narrative gets there. I would note that revenue does not really increase linearly with the number of satellites you launch for various reasons (e.g., limitations on geosynchronous orbital slots, which are quite independent of launch costs), so although lower launch prices will lead to more launches, it probably is not so much as you suppose.
Thanks, for PDF link, e of pi
good points, Shevek23
There is threat called Alternate space shuttles here in this forum. https://www.alternatehistory.com/forum/threads/alternate-space-shuttles.425024/
you might comment in that discussion.
good points, Shevek23
There is threat called Alternate space shuttles here in this forum. https://www.alternatehistory.com/forum/threads/alternate-space-shuttles.425024/
you might comment in that discussion.
Chapter 12: Lofted
“We recommend that: The NASA Modified Launch Services Agreement be extended, as space operations grow, to include interorbit transport services, base camp support services, and other services as appropriate.”
--Pioneering the Space Frontier, 1986
Chapter 12: Lofted--Pioneering the Space Frontier, 1986
Most of a rocket’s weight at take-off is propellant, and Constitution was no exception. As she rose from the launch pad, she burned tons of propellant per second. Bound as she was by the same Newtonian physics that governed all cosmically slow bodies, her acceleration crept up as she left mass behind in Earth’s atmosphere. The stack that had seemed to crawl off the launch pad was, by the time Constitution released Endeavour and her S-IVC, pushing up into space at 5 Gs.
All of a sudden, that acceleration disappeared. Subtly, inaudibly, Endeavour’s structural members flexed in response to the sudden release of load, aluminum and titanium members shifting like springs. In the cockpit, Fred Haise and Dick Truly could sense none of that; only the accelerometers on their dashboards and their own sudden weightlessness confirmed the shout of “MECO” that came over the radio from Constitution.
Then the S-IVC’s engine lit, a feeble successor to Constitution’s five monstrous motors. The Shuttle and its long upper stage began to accelerate in turn, starting at only a third of a gravity. The feeling of weight was not confined to the Orbiter--as Constitution caught some of the gas the J-2S-2 scattered, her crew too felt a shred of the engine’s force.
“Almost feels like the Moon,” observed John Young over the comm loop.
“I don’t know about the Moon, but if you’re done catching our wake, we’ll see you on the other side of the sky,” answered Haise wryly.
One third of a G is not a spectacular acceleration. Had the Lifter not already given her a large vertical velocity, Endeavour would in fact have begun falling back to Earth. Her trajectory was heavily lofted, allowing the S-IVC to burn toward the horizon, to give her a downrange velocity while she ate away at the altitude with which the Lifter had invested her. But with time, it adds up. Slowly but steadily, the propellant burnt off and the accelerometers picked up. Endeavour was on her own path, gathering velocity and altitude by the second, one which would take her far away from the Lifter Constitution, far beyond any Lifter’s capabilities. Men had flown here before, but Endeavour brought a new capability, and would allow them to take small steps and giant leaps of which the first astronauts could only dream.
As the US Army’s frontier forts had once paved the way for settlement of the American West, as the expansion of military aviation paved the way for the explosion of civil jet travel after the Second World War, so NASA, by creating the Lifter, had expanded the market for satellites and other payloads in Earth Orbit--a market that was swiftly filled by satellite television, advanced communications satellites, and, lately, commercial interest in Spacelab and new earth observation systems. By 1989, Martin-Marietta’s ambitions in the field of space launch had become an open secret, and communications giant Motorola was in the planning stages for a new constellation of low-orbiting satellites. At NASA, the sentiment prevailed that the job in Low Earth Orbit was nearing completion, and that it was time to look further outward. As the National Commission on Space had written in 1986, in its report Pioneering the Space Frontier, the purpose of the American civil space program was to “lead the exploration and development of the space frontier, advancing science, technology, and enterprise, and building institutions and systems that make accessible vast new resources and support human settlements beyond Earth orbit, from the highlands of the Moon to the plains of Mars.”
The election of George H. W. Bush and his inauguration in 1989 provided a fertile ground for that new direction. Like Spiro Agnew before him, Bush, as Vice President, had been tasked with overseeing aspects of the American civil space program in President Reagan’s place, reporting directly to Reagan when needed. In 1988, after the loss of Magellan and during the consequent stand-down of the Space Transportation System, Bush met with Administrator James Beggs to discuss the Lifter’s return-to-flight and the Complementary Expendable Launch Vehicle program’s performance in the meantime. During this time, Beggs and Bush also discussed the recommendations of the National Commission on Space and how they could be implemented in the future, after Lifter’s return-to-flight. While under no illusion that he would retain his post as Administrator in 1989 (he already planned to submit the customary resignation to the new President), Administrator Beggs demonstrated an admirable devotion to duty in his last months, trying to make his successor’s job and that of the incoming President as easy as possible.
Authorized by Congressional mandate in 1984, the NCS reflected the growing prominence of planetary exploration and even settlement at NASA. Since the Case for Mars conferences in the 1970s, themselves spurred by the success of the Viking program, the Red Planet, so long viewed as a barren, cratered wasteland, had gotten massively better PR, with ample discussion of the planet’s vast deposits of ice, its tenuous but useful atmosphere, and the tantalizing possibility of finding microscopic alien life. As Werner von Braun’s work with Walt Disney and Willy Ley’s articles in Collier’s had done thirty years earlier, the conferences, and the mass-media articles they generated, had built up some institutional momentum for human exploration of Mars. When the NCS published Pioneering the Space Frontier in 1986, it unambiguously named a human mission to Mars at some undefined future date the nominal goal for America’s civil space program. To accomplish this goal, the NCS called for the development of new, lower-cost launch technology (and an expansion of NASA’s commercial launch contracts beyond Low Earth Orbit), advanced interplanetary propulsion technology, nuclear reactors for in-space power, closed-loop life support systems, and a fully-reusable interorbital tug. It was these recommendations that would inform NASA’s plans for space exploration in the 1990s.
Such ambitious plans, while promising a path forward for NASA to use Lifter to access the Moon and beyond, would have to percolate at the highest levels. In the near term, the larger concern was seeing Lifter on a safe return to flight. Though the initial causes of the STS-116 ignition failure were traced within months, the quality control and procedural changes necessary to address the deeper roots of the issue lasted longer. The time was also needed for the final tests and qualification of the Dual Engine Upper Stage, which added a second J-2S-2 for reduced gravity losses, increased payload, and better performance in abort scenarios. One of the key recommendations of the Magellan Review Board was for any critical missions in the future to make use of the S-IV-D Dual Engine Upper Stage. Even if the extra capacity to orbit wasn’t required, the increased system redundancy in the only part of the STS which was not capable of post-flight inspection and requalification was worth the price. STC’s contract office quickly saw many of its customers making the same decision, seeking to switch existing contracts onto the Space Lifter with DEUS. Even from the beginning, STC had expressed some internal concerns about the increased overhead and process complexity of two stage production lines. The new thrust structure would necessarily contain very few common parts, thanks to changes in feed lines, control runs, stage attitude control, as well as the simple mechanical attachment of the engines. With more and more interest in the S-IV-D and concerns about the increased scrutiny which would be required on future SEUS launches, STC made the bold announcement that they would voluntarily commit to retire the S-IVC entirely, switching existing bookings onto the S-IV-D at cost. It was decision driven not just by customer and public relations: shrewd studies had shown with the increased rates of requests for DEUS launches, much of the cost increases caused by the extra engine were matched by retaining a single common stage for all launches. The date for Space Lifter’s return to flight would be delayed until S-IVD was ready, but the time would allow a clean transition of McDonnell’s production operations to the new stage design.
The ripples from the loss of Magellan were not confined solely to the launch vehicle. The probe’s destruction before reaching orbit had been an inauspicious start to the Mariner Mark II program. Though the spacecraft itself had not been to blame, or even officially part of the program, Magellan was supposed to prove the concept of reusing standardized spacecraft parts to reduce overall mission costs, allowing NASA to launch more spacecraft to more destinations without a radical increase in its budget. Its loss set the program back greatly and left Principal Investigators at universities and laboratories across the US scrambling to preserve their chosen programs.
At the time of Magellan ’s loss, the Mariner Mark II program had converged on a standardized spacecraft bus design, using elements derived from Voyager and Galileo hardware, together with gyroscopes derived from those used on the latest long-lifetime communications satellites. No fewer than three missions had been planned to use the Mariner Mark II chassis--the Comet Rendezvous/Asteroid Flyby mission, the Saturn Orbiter/Titan Probe, and a Neptune Orbiter and Probe. Other missions also called for using the Mariner Mark II design, but were less well-defined and had not begun development. In other words, Mariner Mark II represented NASA’s entire outer-solar-system exploration plans for the next twenty years or more. The Jet Propulsion Laboratory’s scientists and their colleagues elsewhere in the country had staked a lot on the program, and came together to ensure that the program persisted through the doldrums of the Magellan investigation, in preparation for Lifter’s eventual return-to-flight.
As a new Administration came to power in Washington, all the agency’s existing programs came up for review to determine how well they fit into the overall vision, and whether their budgets could be sustained in the coming decade. Though Magellan had never completed her mission, the defenders of the Mariner Mark II program could point to the craft’s well-documented cost savings during construction to defend their program’s claim to similarly reduce costs through the use of standardized components. Magellan had cost far less to construct than Galileo, after all, and advocates for the various Mariner Mark II missions could each point to that success when projecting costs and budget overruns for their projects. With the new emphasis on bold exploration and trailblazing at NASA with the rise of the Space Exploration Initiative, the case for a bold new fleet of planetary probes was well-received at the agency’s headquarters and, when it came to their attention, at the National Space Council.
It helped the program’s case that almost every Mariner Mark II mission had significant European investment, making their cancellation (and subsequent alienation of America’s allies) less attractive to congressmen hungry for their slice of the Peace Dividend (though, as the International Solar Polar Mission had shown just a few years earlier, that approach was not foolproof). The Saturn Orbiter/Titan Probe mission, for example, had begun life in 1982 as a European Science Foundation study into possible joint missions with the Americans, before being adopted by NASA (which had been looking into Saturn missions since the 1970s) as a primary science objective in 1983. The Neptune Orbiter shared a lot of the SO/TP instruments, and in the 1980s was essentially an appendage to that program. The Comet Rendezvous/Asteroid Flyby probe included a number of European instruments, many of them spares from the 1986 Halley Armada, and a set of European-designed penetrator-landers. One by one, the Mariner Mark II programs found their way into NASA’s budget authorizations, and began to take physical shape.
In the meantime, NASA was looking to polish its image as the Magellan incident report’s conclusions were taken to heart. During the initial months after Magellan ’s failure, the press had been filled with stories criticizing NASA, and for many unengaged by spaceflight it was the first time they had thought heavily of NASA in almost a decade. With the entire Lifter and Shuttle fleet stood down, NASA was taking the time to give each vehicle an intensive inspection and overhaul, but even so there were more vehicles than NASA had inspection bays. NASA’s public affairs office decided to combine the two facts to take advantage of the interest, and refocus it more positively. Shortly after the Magellan Review Board reported, and with the STS on the road to return to flight, NASA announced that the STS fleet would be making appearances at a variety of airshows during the summer 1988 series. The Lifters were capable of ferrying themselves to any runway capable of handling a 747, while the Shuttles were carried routinely on the backs of specially modified 747s. However, except for a few publicity events shortly after the debut of the system, these capacities had only been used to ferry Lifters and Shuttles back and forth across the Gulf Coast and the Southwest, swapping between Vandenberg and Florida or returning for inspections. By the end of the year, every major airshow in the United States had been visited by a Lifter and Shuttle. Plans were even considered to fly the Lifter internationally, carefully working its way north and east across the Atlantic to make an appearance at the Farnborough Air Show. Ultimately, the logistics and time required meant that only the Space Shuttle Destiny (which was light enough its 747 carrier could still make the trans-Atlantic trip uninterrupted) was able to visit--once again, the Shuttle would go where Lifter could not follow. The Space Shuttle and Space Lifter were stars of the 1988 air show circuit, resulting in endless home videos and Polaroid images of Lifters making low flyovers or press footage of crowds circulating around grounded spacecraft.
While NASA’s image was being rebuilt, however, concerns floated around various space agencies about the length of the stand down. Shuttle’s delays in launches were fortunately not critical: Discovery had carried the Long Duration Exposure Facility back from its sixth orbital stint as part of a satellite deployment mission on STS-112 in February of 1988. Spacelab’s orbital status was more of a concern, as it had only sufficient propellant aboard for eighteen months of independent orbital stabilization. Fortunately, the Review Board’s verdict seemed to indicate that the Shuttle should be flying again well before the deadline, and the station’s man-tended design meant it was robust enough to last almost a year between visits. Nevertheless, with no way to actively address any developing situations, many engineers in Houston sweated long hours over any signs of potentially debilitating issues aboard the platform. The USAF had already planned to delay major servicing of the LUCID spy satellites until the DEUS could enhance payload to their eccentric orbits, so the largest delay was to NASA’s own Hubble Space Telescope. After more than a decade in incubation, the telescope was finally scheduled for launch in late 1988, and the delay directly impacted the telescope’s launch schedule. More than one program manager within Hubble breathed a sigh of relief that their delays had prevented them from being assigned a slot nearer STS-116, and the flagship observatory was specifically cited in the Magellan Report as an example of a payload which should receive the additional redundancy of DEUS going forward even though its mass didn’t require it.
Waiting with Hubble for a launch assignment after the return to flight were more than a dozen large satellites, ranging from military signals intelligence to commercial satellite platforms. The operators of those under three tons were able to consider arranging flights with ESA’s Ariane, though manufacturing delays on the little-flown vehicle left it unable to rapidly meet the spike of demand. However, for those over 4 tons, which were beginning to approach half of the global commercial satellite manifest, there was only one alternative to waiting out the Lifter’s stand-down: Titan. Though the CELV contract called for Titan to be ready for a launch on six months notice, in fact activating the Titan launch site at Cape Canaveral alone took more than eight, by which time the issues with SEUS had been exposed, the decision to switch all flights to DEUS made, and dates for the return of the Lifter were being discussed. Still, with additional concerns about the potential for further delays in the new stage’s introduction circulating, the first launch of Titan in more than three years went ahead. On October 15, 1988, a Titan 3D roared off the pad at LC-41, carrying a classified military payload to orbit. Two more would fly in December and January to relieve national defense backlogs during the stand down, but the major development came from joint lobbying from Martin-Marietta and a broad group of satellite builders to offer some of the remaining 27 stockpiled Titans at launch cost for commercial customers who had booked Space Lifter flights during the stand-down. For these operators, the sight of launchers capable of carrying their payloads flying while they watched loan payments and stock prices fluctuate was frustrating, as there had been a common impression that STC was, in some sense, backed by the full faith and credit of NASA and the USAF. With that confidence shaken, many were eager for any alternative, and the USAF’s reluctance to release Titans at any price lead to them being seen as a dog in the manger. For the USAF’s part, there were concerns that releasing some or all of the Titan stockpile during the current stand-down could set a precedent for future contingencies, and deplete a reserve which was now seen as having proved its worth.
While Martin played the role of the business which wanted to offer a product but was restrained by the USAF, their real goals were more complex. While they could use the extra income which came from each launch of a Titan from the stockpile to help fund their Trans-Pacific Launch Industries venture with Mitsubishi, they were just as unwilling to actively antagonize the Department of Defense, and they were careful to always leave satellite builders the ones most loudly calling for the release of Titans. Instead, their complaints about restrictions on commercial sales from the Titan stockpiles were an excuse to rub the industry’s nose in STC’s failure, and drive home the benefits of a Titan-class alternative to augment and (if necessary) substitute for Space Lifter. After all, TPLI’s own new launcher was designed to address just that segment, and the demands now laid further groundwork for sales of its services later.
Among public relations outreach and launch schedule jockeying on the ground, STC finally made major steps towards Lifter’s return to flight in the fall. The first Dual-Engine Upper Stage, SIV-D-T, was hot fired at Stennis test site on October 15, 1988, marking a critical step for Lifter to return to the skies. In spite of the pressure placed on the tests, or more accurately because of them, the DEUS test program was cautious and incremental, seeking to ensure that the new thrust structure and the systems for igniting and controlling the twin engines would be more reliable than their flight-proven single-engine equivalents. As test engineers rang in the new year, the earliest expected date for the Lifter’s return to service slipped from February 1989 into March. The test program’s delays were frustrating to those depending on Space Lifter for their rides to orbit, but NASA was determined that the lax safety culture which had contributed to Magellan’s loss would not be allowed to reemerge. The S-IV-D would not debut until it was fully qualified, even as the stand down stretched closer and closer to a full year. While the Lifter had been off the flight line, though, decisions were being discussed which would shape the future of the space program.
The idea of double SIVD stages (the upper one being a dry workshop, space station, Mars transport habitation module (use two, rotating at the end of a longish arm)) is drool-worthy.