Evolution vs. Revolution: The 1970s Battle for NASA's Future

Sunlight glints off NASA Marshall Space Flight Center's proposed Power Module in this artist concept by Junior Miranda.
According to historians Andrew Dunar and Stephen Waring, writing in their 1999 NASA-funded history Power to Explore: A History of Marshall Space Flight Center, in the 1970s two lines of thought emerged within NASA concerning manned spaceflight's course after the Space Shuttle became operational. On the one hand, there was the "revolutionary" line taken by Johnson Space Center (JSC) in Houston, Texas. On the other was the "evolutionary" line of NASA Marshall Space Flight Center (MSFC) in Huntsville, Alabama.

At JSC, many managers assumed that, as soon as the Shuttle became operational, NASA would get a green light to assemble a large, new-design, multipurpose Space Station in low-Earth orbit (LEO). They envisioned that a 1980s President would make a speech much like President John F. Kennedy's 25 May 1961 “moon speech.” Visionary goal thus proclaimed, the funding floodgates would open.

At MSFC, by contrast, many managers expected that NASA budgets would remain tight for the foreseeable future, so that any space technology development that took place would need to be incremental; that is, it would have to begin with existing space hardware and occur in small steps. MSFC's work on the Skylab Orbital Workshop, a temporary LEO space station launched in May 1973 on the last Saturn V rocket to fly, probably helped to shape their outlook.

The 169,950-pound Skylab "cluster," which comprised the Multiple Docking Adapter, the Apollo Telescope Mount (ATM), and the Orbital Workshop, had been conceived originally as an element of the Apollo Applications Program (AAP). As its name implies, AAP had been meant to apply hardware developed for the Apollo lunar program to new tasks. The Skylab Orbital Workshop was a converted Saturn S-IVB stage outfitted with experiment apparatus, crew quarters, and supplies for visiting three-man crews. Three crews were launched to Skylab in 1973-1974; the last orbited the Earth for 84 days.

The Skylab Orbital Workshop floats serenely over the Earth, but this image bears evidence of its nearly disastrous launch and the heroic efforts that saved it. Skylab's reflective meteoroid shield deployed during ascent and peeled away, tangling one of its wing-like solar arrays in debris and loosening the other. A stage separation rocket motor then blasted away the loose array and the tangled array refused to open. Skylab was starved for electricity while temperatures inside it soared, threatening to spoil food, film, and medicines. The first Skylab crew (Charles Conrad, Paul Weitz, and Joseph Kerwin) deployed a sun shield and forced open the jammed array. Skylab went on to host astronauts for a total of 171 days. Image credit: NASA.
NASA built most of a second Skylab, but was unable to secure funding to complete it, launch it into orbit, and launch crews to it. The first Skylab was a success, so MSFC might have expected on that basis to have "earned" funding for the second. The Huntsville Center had, however, learned during the 1960s not to equate success with rewards. It had been responsible for the Saturn V moon rocket, the largest and most powerful launcher ever built. Even as MSFC succeeded in making the mighty Saturn V work, however, it began to suffer funding and staff cuts that by the time Skylab flew would make it a shadow of its former self.

When MSFC engineers looked at the Space Transportation System (STS), as NASA called the Space Shuttle and its stable of expendable upper stages and European-built Spacelab components, they saw not the promise of a big new space station, but rather a system which, once operational, could benefit from evolutionary development. In particular, they noted that Spacelab, which MSFC was assigned to integrate with the Shuttle, could not reach its potential as an orbiting laboratory while the Shuttle Orbiter's planned maximum time in space was only seven days. The Orbiter and its payloads would rely for electricity on the former's fuel cells, which meant that the quantity of fuel-cell reactants the Orbiter could carry would determine their endurance.

The Space Shuttle Orbiter with its Payload Bay doors open to space. A drum-shaped, European-built Spacelab module is shown as a cutaway.  Curved panels raised above the front half of the doors are radiators. The Spacelab module is located near the rear of the Payload Bay to ensure that the Orbiter's center of gravity is placed properly for maneuvers and landing. Image credit: NASA.
In early 1977, with the first STS flight test officially planned for March 1979, MSFC proposed "the first step beyond the baseline STS" — a Power Module (PM) capable of supplying 25 kilowatts of electricity continuously. The PM was partly inspired by joint Department of Energy/NASA Solar Power Satellite studies of the 1970s.

The solar-powered PM was meant to be deployed into LEO from a Shuttle Orbiter payload bay and left in space for up to five years. A succession of Orbiters bearing Spacelab modules and pallets in their payload bays would dock with the PM and use its electricity to remain in orbit for up to 30 days at a stretch.

Alternately, a Shuttle Orbiter could attach a "freeflyer" payload to the orbiting PM and leave it to operate on its own. This appealed to materials scientists, who worried that astronauts' movements on board the Shuttle Orbiter and Spacelab would rattle and ruin their microgravity experiments. Orbiters would periodically dock with the materials science freeflyer/PM combination to retrieve experiment products — for example, large flawless crystals — and replenish raw materials.

In addition to electricity, the PM "building block" would provide thermal and attitude control. The latter would permit a docked Orbiter to conserve its Reaction Control System propellants. Freeflyer payloads meant to be docked with the PM could be built without thermal and attitude control systems, reducing their cost.

Image credit: NASA.
MSFC engineers planned at first to base the PM on the Skylab ATM design. They quickly found, however, that modifying the ATM to meet stringent Orbiter payload bay safety requirements would cost more than a new design. They retained the ATM's octagonal cross-section, however, because they found that it made efficient use of the Orbiter's cylindrical payload bay volume while providing flat surfaces upon which to mount subsystems.

Although it nixed the ATM-based design, MSFC still aimed to lower the PM's cost by using subsystems developed for Skylab, Spacelab, Shuttle, and other programs. These included three Skylab Control Moment Gyros for attitude control and four curved Shuttle Payload Bay door radiators for thermal control. MSFC planned to update and improve Skylab systems used in the PM based on Skylab flight experience. All major PM subsystems would be redesigned for easy replacement by spacewalking astronauts.

The 31,000-pound PM would measure 55 feet long from the framework holding its aft- and side-facing international docking ports to the forward ends of its stowed twin solar arrays. This would leave room in the Shuttle Orbiter's 15-by-60-foot Payload Bay only for a docking tunnel with an international docking port. The tunnel would be bolted to the forward wall of the bay over the hatch linking the bay to the Shuttle Crew Compartment.

This NASA artwork shows a Space Shuttle Orbiter bearing a Spacelab module in its Payload Bay docked with a separately launched Power Module which extends forward over the Orbiter Crew Compartment.
Upon arrival in LEO, the astronauts would open the Shuttle Orbiter's Payload Bay doors and release the five pins that secured the PM in the bay. They would then use the Orbiter's robot arm to lift the PM from the bay and berth its side-facing docking port on the Orbiter docking port. This would position the module so that it extended out over the Crew Compartment.

The astronauts would next extend the PM's twin solar arrays. Fully extended, each wing-like array would measure 131 feet long by 30 feet wide. They would together span a little more than 276 feet. MSFC sized the arrays to generate a total of 59 kilowatts of electricity; that is, 34 kilowatts more than the PM would supply to Spacelab-carrying Orbiters and freeflyers. A portion of this excess would power PM systems, but the majority would charge batteries in the PM so that it could supply a constant 25 kilowatts throughout its roughly 90-minute orbital day-night cycle.

Close-up of Power Module showing international docking ports and curved radiator panels. Image credit: Junior Miranda.
MSFC acknowledged that the big solar arrays would degrade over time; its engineers estimated that over five years they would lose 5% of their generating capacity. Similarly, the PM's batteries would gradually lose their ability to charge and discharge. After five years, a Shuttle Orbiter might be sent up to recover the PM and return it to Earth for refurbishment. Another Orbiter would then launch it back to LEO to continue its duties.

MSFC engineers presented the PM concept to scientists at an MSFC-sponsored solar-terrestrial physics workshop in October 1977. They found broad support for the new capabilities the PM would give to the baseline STS.

Lots of living space: Skylab, Power Module, Spacelab-based add-on supply module, Shuttle Orbiter, and Payload Bay-mounted Spacelab module. Image credit: Junior Miranda.
This view emphasizes the solar arrays on the Power Module and Skylab. The 276-foot span of the Power Module arrays dwarfs the Shuttle and Skylab. The Skylab "wing" array lost during launch in May 1973 is conspicuous by its absence; also notable are two Apollo Telescope Mount "windmill" solar arrays stowed to make way for the Power Module and Orbiter. Image credit: Junior Miranda.
They also proposed that the PM become part of NASA plans to reuse Skylab. MSFC contractor McDonnell Douglas had "interrogated" the abandoned Orbital Workshop's data handling system and found that, nearly four years after its last crew had returned to Earth, reactivation remained feasible. The first step toward Skylab reuse would be for a Space Shuttle to rendezvous with it late in 1979 and boost it to a longer-lived orbit.

The PM would be a late addition to the revitalized Skylab cluster; MSFC did not expect that the new STS element would reach LEO for the first time until 1983, by which time several Shuttle Orbiters would already have visited Skylab. Once added to Skylab, however, the PM would enable Skylab to support as many as six astronauts without a Shuttle Orbiter present. They would perform experiments with large-scale space construction and early space industrialization.

MSFC engineers hoped that the PM might also contribute toward NASA's quest for Skylab's successor. They envisioned that PMs attached to Shuttle Orbiters, free-flyers, and Skylab might lead to PMs attached to Spacelab-derived habitat and laboratory modules during the 1980s: in other words, a new NASA Space Station.

In 1978, the Huntsville center contracted with Lockheed Missiles and Space Company to study PM evolution. MSFC expected that PM development might lead to simultaneous operation of several small specialized "space platforms," each with at least one PM attached. The platforms would not need to be staffed continuously. MSFC argued that several small platforms would best serve scientific and engineering disciplines with conflicting needs, and might cost less than a single large station besides.

In early 1979, NASA Headquarters authorized MSFC to spend $90 million on PM hardware development. The Huntsville center created a PM Project Office in March 1979. At about the same time, however, the space agency abandoned plans to reuse Skylab because the Space Shuttle would not be ready in time to prevent its uncontrolled reentry. Skylab reentered Earth's atmosphere over Australia on 11 July 1979.

JSC, meanwhile, pitched a new-design Space Operations Center (SOC). The space station would include hangars for reusable auxiliary spacecraft and satellite repair, robot arms, habitat and laboratory modules, and truss-mounted solar arrays spanning more than 400 feet. It was conceived primarily as a "space shipyard," a role inspired partly by JSC's 1970s enthusiasm for Solar Power Satellites.

Artist concept of the module cluster of the Space Operations Center (SOC). Most modules are a little less than 60 feet long by 15 feet wide (the length and width of the Space Shuttle Payload Bay). At lower left is a "false Payload Bay" for satellite servicing and spacecraft assembly. Had the SOC been built, this would have included robot arms. A Service Module partly covered with gold thermal blankets is located at upper right and a hexagonal hangar is located below it. The artist has included a Spacelab-derived module near a Shuttle docking port at left. Image credit: NASA.
STS-1, the maiden flight of Columbia, the first Space Shuttle Orbiter, took place in April 1981. James Beggs, President Ronald Reagan's choice for NASA Administrator, was confirmed two months later. Beggs soon sought presidential approval for a Space Station. This move seemed to favor JSC's revolutionary vision. At the same time, however, Beggs informed MSFC that he wanted to buy the new station "by the yard" – that is, as money became available. This approach seemed more in line with MSFC thinking.

In November 1981, NASA Headquarters halted PM, SOC, and other station-related work at MSFC and JSC. According to Dunar and Waring, it did this to take charge of station development and to end MSFC-JSC rivalry. Following Reagan's January 1984 State of the Union Address, in which he called upon NASA to build a Space Station by 1994, JSC's revolutionary vision seemed to win out. JSC was designated "lead center" for Space Station in early February 1984.

Although Reagan authorized NASA to spend only the $8 billion Beggs had told him the Space Station would cost and had specifically called for a space laboratory in his State of the Union Address, the agency's first baseline station design, the "Dual Keel," was an elaborate combination of lab, Earth/space observatory, and shipyard measuring more than 500 feet wide. Like the SOC, it included a small fleet of freeflyers and auxiliary vehicles. It also included a pair of solar-dynamic power systems — a NASA Lewis Research Center innovation — for generating large amounts of electricity.

The Dual-Keel Space Station design unveiled shortly after the January 1986 Challenger accident was dead on arrival, though NASA sought to ensure a future for the design until 1990. Image credit: NASA.
The Dual Keel's complex multipurpose design immediately came in for criticism. Materials scientists, for example, complained that space construction, the comings and goings of auxiliary spacecraft, the whirling turbines of solar-dynamic power systems, the presence of a large crew, and atmospheric drag on such a large structure were bound to spoil the station's microgravity research environment. Congress, meanwhile, accused NASA of low-balling its cost estimate to gain the project's approval.

Congressional cost containment, combined with the 28 January 1986 Challenger accident, concern over the number of assembly and maintenance spacewalks the station would need, and a rapidly expanding U.S.-Russian space partnership (one which would have been unthinkable when Reagan delivered his January 1984 speech), led to a decade-long series of station redesigns. The Space Station shrank and lost many of its proposed capabilities. This untidy evolution yielded the International Space Station (ISS), a U.S.-Russian hybrid with Japanese and European labs and Canadian robotics.

Early days of the International Space Station: from upper left to lower right are visible a Progress freighter, the Service Module with docking node, the FGB, and U.S. Node 1. Image credit: NASA.
Ironically, the first ISS element launched into space amounted to a Power Module. The Russian-built, Russian-launched, U.S.-funded FGB provided the second ISS element to reach space, U.S. Node 1, with electricity and attitude control from December 1998 to July 2000, when they were joined by a mini-space station – the Russian-built, Russian-launched Service Module, which had originally been intended as the "base block" of the Soviet Union's Mir-2 station. At that point, ISS became capable of supporting long-duration crews.


Guntersville Workshop on Solar-Terrestrial Studies, NASA Conference Publication 2037, "summary papers from a University of Alabama in Huntsville/NASA Workshop conducted 13-17 October 1977, at Lake Guntersville State Park Convention Center, Guntersville, Alabama," NASA George C. Marshall Space Flight Center, 1978.

"The 25 kW Power Module – First step beyond the baseline STS," G, Mordan; paper presented at the American Institute of Aeronautics and Astronautics Conference on Large Space Platforms: Future Needs and Capabilities, held in Los Angeles, California, September 1978.

25 kW Power Module Updated Baseline System, NASA TM-78212, NASA George C. Marshall Space Flight Center, Huntsville, Alabama, December 1978.

Power to Explore: a History of Marshall Space Flight Center, 1960-1990, NASA-SP-4313, Andrew J. Dunar and Stephen P. Waring, NASA History Office, 1999.

More Information

What Shuttle Should Have Been: NASA October 1977 Space Shuttle Flight Manifest

McDonnell Douglas Phase B Space Station (1970)

Where to Launch and Land the Space Shuttle? (1971-1972)


  1. Ah, the things that could have been. That is why I love your blog! At least Revel gave us a Space Operation Center model kit!

    1. I bought several of those. NASA was talking about using Space Station modules as lunar base modules, so I built several lunar outposts from those kits.


  2. I never realized that the shuttle was that limited by lack of long-duration power generation. Reading this makes sense of why shuttle flights never lasted all that long. I remember hearing about he power module a long time ago and did wonder where it went. Also explains why the Spacelab missions were so short in comparison to what the Russians were accomplishing at the same time. It's also interesting how so many hopes and dreams crashed with Skylab while the shuttle was eating up any hopes of saving it.

  3. The Shuttle was meant to convey a crew and supplies to a Space Station; it only ever was used as a laboratory because no one wanted to pay for a U.S. Space Station. So, it wasn't a very good laboratory - one might say experiments performed on board only hinted at what might be feasible on board a long-term Space Station. Over time, however, we upped its endurance, so that the STS-107 crew, for example, was returning from a three-week mission on board Columbia, the oldest Orbiter, when they were killed. I've not calculated the average mission duration later in the Shuttle program, but I have the impression it was closer to two weeks than one week.


  4. This is what befuddles me with the US government thinking in space:
    1. Several plans for a space-station/orbiting laboratory but most went unrealized.
    2. Went to the Moon, but so what, no further development.
    3. A very cool space plane, but many changing, if not confusing mission purposes. (OK, I know it was oversold relative to its capabilities, so it makes sense).

    Alternatively, the Soviets also work in space and went to the Moon robotically.
    1. Curious why the Soviets/Russians never slowly and steadily developed a Moon (robotic if nothing else) presence, like they slowly and steadily developed space stations?
    2. Seems like the competition thing would have taken a different turn with Hammer and Sickle robots doing proof of concept mining, testing, developing on the Moon through the 70s/80s.

    The above is why the whole (manned journey to) Mars idea seems a long ways off. It's a big leap technically and logistically - meaning lots of money. NASA can plan for the 2030s, but the funding has to be there. And the commitment. Have not seen either. And with the wisdom of defunding commercial crew to send money to Roskosmos, face-palm.

  5. Ben:

    We have very seldom done space stuff because space. We've nearly always done it because of the Cold War, pork distribution, the quest for intl. prestige, the desire to foster a body of scientists and engineers whose experience and talents could be tapped in time of war, etc., etc. Add the fact that few people really *get* space - and that not infrequently includes people who profess an interest in it, sadly - and all sorts of dumb stuff happens.

    Of course, this is not all that different from many other spheres of human activity. As someone once said (it's attributed to several people) "history is just one damned thing after another."

    Commercial crew is to me one recent manifestations of this. I think it is in the same league as Shuttle without a Station as far as goofy space decisions goes. It's an attempt to get a space station access capability on the cheap. Which is dumb when it comes to spaceflight. Spaceflight is hard - if you don't spend enough to make your systems robust. Plus, what is it about Commercial Crew that excites people? Our expectations have shrunk so far that launching an Espresso machine and harvesting plants (not for the first time, by the way) gets a lot of attention. F*** that.

    Unfortunately, the past Administration painted us into a corner by canceling the Shuttle and not adequately funding its replacement and the current Administration isn't all that interested in a sound and coherent space policy. So, if we want to access "our" station, we have to pay the Russians.

    I hope that Orion will end up as a cislunar spacecraft, and that includes crew/cargo missions to a space station - though not necessarily ISS, and not necessarily in LEO. It's not really a step toward Mars any more than Dragon is. The neat thing about Orion and SLS is we can get back to Skylab-type stations if we want, and that's a thing we should want. I could see us putting these in several different locations in cislunar space.

    By the way, I deleted your other comment. If you would like to know why, please send me an email.


    1. "It's an attempt to get a space station access capability on the cheap."

      It is, but that's because NASA wouldn't be able to afford to do it on its existing budget (not without cutting something else), and it would take much longer. As it is, Orion will end up costing something around $18 billion for development - and take 12 years to develop and build (perhaps longer if the ESA service module is delayed, as it might be). You can't do space flight on the cheap, but it is fair to ask if it has to be as expensive as it is. Must we always use cost plus contracts, for example? The reality is that NASA funding has been stagnant in real dollars for a couple decades now, and there is no prospect of that changing. NASA has to do whatever it is going to do on a fixed budget. We all wish it were otherwise, but that is the political reality. The support just isn't there to spend more.

      I think it's fair to ding the Bush Administration for whacking the STS without a successor system in place and a commitment to get it ready for use before the Shuttle was retired. But that was always going to be a difficult prospect, because every Shuttle mission after the return to flight post-Columbia represented a significant risk of crew loss.

  6. We don't need a new Skylab, the ISS is a much better long duration Space Station.
    The ISS did not face any major hazards like the MIR did:
    - A fire inside did destroy unknown content.
    - The module Spektr was hit by a Progress transporter.

    The ISS was so expensive to build, let's support this base above for more than the next fifteen years. More we will not get over that time frame. Manned Mars will be beyond 2050, maybe more.

    1. Clearly agree on Manned Mars. "It's always 20 years away" really holds true. Indulge me my analysis:
      1. Federal space budget is not oriented toward that level of expense.
      2. No plan or commitment to a plan. (Lots of tech development in play though)
      3. Administrations and goals change in 4-year cycles.
      4. Commercial efforts do not have the infrastructure (or payback) for the foreseeable (20-40 years) future.
      5. Budgetary constraints and pressures will increase (debt/entitlements in federal budget), not decrease.

      So I suggest Mars will have human boot prints on it in 2120. Because, you know, in 2100, we'll be saying "Humans on Mars is only 20 years away."

      Next 20 years from now science is on track for an exoplanet astronomy bonanza and a manned exploration dark age.

  7. I think one significant point of the MSFC scenario is that a single station can't serve any one purpose well; to accomplish the best results across the board one needs multiple automated platforms and specialized stations in a variety of locations. These stations can exist at the same time or can occur in a logical series. Personally, I do not think that having paid an excessive cost is a good justification for continuing to do something. That said, I agree that we need not replace ISS immediately; and anyway, we cannot, for SLS and Orion aren't ready.


  8. I recall when you ran the first version of this at the old place, and I still think it's one of the most eye-opening - and depressing - pieces of space program history that you've done.

    I think there's a consensus now that the STS was a misguided program, however marvelous its capabilities were in certain ways. But had MSFC had its way, we could have gotten a good deal more bang for that buck over its 30 year history.

  9. NASA JSC has long exerted a powerful influence on NASA; when I worked there, the consensus seemed to be that it *was* NASA and that no one would give a damn about space at all if it weren't for JSC's piloted missions. The belief was strong that everyone shared JSC's view of itself.

    I have a real fondness for the place and many of its people. That said, I believe that the evidence is strong that its influence on policy has not in general been positive. For example, old-timers at JSC were proud of what they called "the give-away," when they handed off AAP/Skylab to MSFC, which they had worked since 1966 to turn into a dead-end, and turned their efforts toward Shuttle/Tug/Station.

    There are many other examples: to note just one, JSC flouted Code Z exploration planning when it took over the Space Exploration Initiative in 1989. Had it instead built on the work performed at NASA Headquarters in the two or three years prior to Bush the Elder's 20 July 1989 SEI speech, it could have presented to the White House and Congress in November 1989 a far more realistic plan for accomplishing SEI and not provided SEI's many critics with quite so much ammunition. Perhaps something really obvious might have come of SEI then - probably nothing as elaborate as a moon base and a piloted Mars mission, but something. As it was, the main thing that came out of SEI was "faster-better-cheaper," which was not really the disaster many make it out to be. It marked the infusion of SDI technology into the civilian space program.

    The story is, of course, much more complex and nuanced than bad JSC/good MSFC - the latter undermined itself in the 1960s by continually seeking the next big NASA project so it could continue its work on advanced propulsion and giant rockets, even though that just wasn't in the cards. Similarly, JSC and JPL have typically fought each other, and both sides deserve blame for the fumbles that have resulted from the artificial separation of the piloted and robotic programs.

    Now that I've alienated everyone I'll stop. :-)


    1. "The story is, of course, much more complex and nuanced than bad JSC/good MSFC..."

      Just to be clear, I never thought you were making such a claim - only that JSC's role in *this* particular policy debate was unfortunate.


I like hearing from my readers. No rules except the obvious ones - please keep it civil and on topic.

Advertiser comments have led me to enable comment moderation.