White, Rowland. Into the Black. New York: Touchstone, 2016. ISBN 978-1-5011-2362-7.
On April 12, 1981, coincidentally exactly twenty years after Yuri Gagarin became the first man to orbit the Earth in Vostok 1, the United States launched one of the most ambitious and risky manned space flights ever attempted. The flight of Space Shuttle Orbiter Columbia on its first mission, STS-1, would be the first time a manned spacecraft was launched with a crew on its first flight. (All earlier spacecraft were tested in unmanned flights before putting a crew at risk.) It would also be the first manned spacecraft to be powered by solid rocket boosters which, once lit, could not be shut down but had to be allowed to burn out. In addition, it would be the first flight test of the new Space Shuttle Main Engines, the most advanced and high performance rocket engines ever built, which had a record of exploding when tested on the ground. The shuttle would be the first space vehicle to fly back from space using wings and control surfaces to steer to a pinpoint landing. Instead of a one-shot ablative heat shield, the shuttle was covered by fragile silica tiles and reinforced carbon-carbon composite to protect its aluminium structure from reentry heating which, without thermal protection, would melt it in seconds. When returning to Earth, the shuttle would have to maneuver in a hypersonic flight regime in which no vehicle had ever flown before, then transition to supersonic and finally subsonic flight before landing. The crew would not control the shuttle directly, but fly it through redundant flight control computers which had never been tested in flight. Although the orbiter was equipped with ejection seats for the first four test flights, they could only be used in a small part of the flight envelope: for most of the mission everything simply had to work correctly for the ship and crew to return safely. Main engine start—ignition of the solid rocket boosters—and liftoff!

Even before the goal of landing on the Moon had been accomplished, it was apparent to NASA management that no national consensus existed to continue funding a manned space program at the level of Apollo. Indeed, in 1966, NASA's budget reached a peak which, as a fraction of the federal budget, has never been equalled. The Saturn V rocket was ideal for lunar landing missions, but expended each mission, was so expensive to build and operate as to be unaffordable for suggested follow-on missions. After building fifteen Saturn V flight vehicles, only thirteen of which ever flew, Saturn V production was curtailed. With the realisation that the “cost is no object” days of Apollo were at an end, NASA turned its priorities to reducing the cost of space flight, and returned to a concept envisioned by Wernher von Braun in the 1950s: a reusable space ship.

You don't have to be a rocket scientist or rocket engineer to appreciate the advantages of reusability. How much would an airline ticket cost if they threw away the airliner at the end of every flight? If space flight could move to an airline model, where after each mission one simply refueled the ship, performed routine maintenance, and flew again, it might be possible to reduce the cost of delivering payload into space by a factor of ten or more. But flying into space is much more difficult than atmospheric flight. With the technologies and fuels available in the 1960s (and today), it appeared next to impossible to build a launcher which could get to orbit with just a single stage (and even if one managed to accomplish it, its payload would be negligible). That meant any practical design would require a large booster stage and a smaller second stage which would go into orbit, perform the mission, then return.

Initial design concepts envisioned a very large (comparable to a Boeing 747) winged booster to which the orbiter would be attached. At launch, the booster would lift itself and the orbiter from the pad and accelerate to a high velocity and altitude where the orbiter would separate and use its own engines and fuel to continue to orbit. After separation, the booster would fire its engines to boost back toward the launch site, where it would glide to a landing on a runway. At the end of its mission, the orbiter would fire its engines to de-orbit, then reenter the atmosphere and glide to a landing. Everything would be reusable. For the next mission, the booster and orbiter would be re-mated, refuelled, and readied for launch.

Such a design had the promise of dramatically reducing costs and increasing flight rate. But it was evident from the start that such a concept would be very expensive to develop. Two separate manned spacecraft would be required, one (the booster) much larger than any built before, and the second (the orbiter) having to operate in space and survive reentry without discarding components. The orbiter's fuel tanks would be bulky, and make it difficult to find room for the payload and, when empty during reentry, hard to reinforce against the stresses they would encounter. Engineers believed all these challenges could be met with an Apollo era budget, but with no prospect of such funds becoming available, a more modest design was the only alternative.

Over a multitude of design iterations, the now-familiar architecture of the space shuttle emerged as the only one which could meet the mission requirements and fit within the schedule and budget constraints. Gone was the flyback booster, and with it full reusability. Two solid rocket boosters would be used instead, jettisoned when they burned out, to parachute into the ocean and be fished out by boats for refurbishment and reuse. The orbiter would not carry the fuel for its main engines. Instead, it was mounted on the side of a large external fuel tank which, upon reaching orbit, would be discarded and burn up in the atmosphere. Only the orbiter, with its crew and payload, would return to Earth for a runway landing. Each mission would require either new or refurbished solid rocket boosters, a new external fuel tank, and the orbiter.

The mission requirements which drove the design were not those NASA would have chosen for the shuttle were the choice theirs alone. The only way NASA could “sell” the shuttle to the president and congress was to present it as a replacement for all existing expendable launch vehicles. That would assure a flight rate sufficient to achieve the economies of scale required to drive down costs and reduce the cost of launch for military and commercial satellite payloads as well as NASA missions. But that meant the shuttle had to accommodate the large and heavy reconnaissance satellites which had been launched on Titan rockets. This required a huge payload bay (15 feet wide by 59 feet long) and a payload to low Earth orbit of 60,000 pounds. Further Air Force requirements dictated a large cross-range (ability to land at destinations far from the orbital ground track), which in turn required a hot and fast reentry very demanding on the thermal protection system.

The shuttle represented, in a way, the unification of NASA with the Air Force's own manned space ambitions. Ever since the start of the space age, the Air Force sought a way to develop its own manned military space capability. Every time it managed to get a program approved: first Dyna-Soar and then the Manned Orbiting Laboratory, budget considerations and Pentagon politics resulted in its cancellation, orphaning a corps of highly-qualified military astronauts with nothing to fly. Many of these pilots would join the NASA astronaut corps in 1969 and become the backbone of the early shuttle program when they finally began to fly more than a decade later.

All seemed well on board. The main engines shut down. The external fuel tank was jettisoned. Columbia was in orbit. Now weightless, commander John Young and pilot Bob Crippen immediately turned to the flight plan, filled with tasks and tests of the orbiter's systems. One of their first jobs was to open the payload bay doors. The shuttle carried no payload on this first flight, but only when the doors were open could the radiators that cooled the shuttle's systems be deployed. Without the radiators, an emergency return to Earth would be required lest electronics be damaged by overheating. The doors and radiators functioned flawlessly, but with the doors open Young and Crippen saw a disturbing sight. Several of the thermal protection tiles on the pods containing the shuttle's maneuvering engines were missing, apparently lost during the ascent to orbit. Those tiles were there for a reason: without them the heat of reentry could melt the aluminium structure they protected, leading to disaster. They reported the missing tiles to mission control, adding that none of the other tiles they could see from windows in the crew compartment appeared to be missing.

The tiles had been a major headache during development of the shuttle. They had to be custom fabricated, carefully applied by hand, and were prone to falling off for no discernible reason. They were extremely fragile, and could even be damaged by raindrops. Over the years, NASA struggled with these problems, patiently finding and testing solutions to each of them. When STS-1 launched, they were confident the tile problems were behind them. What the crew saw when those payload bay doors opened was the last thing NASA wanted to see. A team was set to analysing the consequences of the missing tiles on the engine pods, and quickly reported back that they should pose no problem to a safe return. The pods were protected from the most severe heating during reentry by the belly of the orbiter, and the small number of missing tiles would not affect the aerodynamics of the orbiter in flight.

But if those tiles were missing, mightn't other tiles also have been lost? In particular, what about those tiles on the underside of the orbiter which bore the brunt of the heating? If some of them were missing, the structure of the shuttle might burn through and the vehicle and crew would be lost. There was no way for the crew to inspect the underside of the orbiter. It couldn't be seen from the crew cabin, and there was no way to conduct an EVA to examine it. Might there be other, shall we say, national technical means, of inspecting the shuttle in orbit? Now STS-1 truly ventured into the black, a story never told until many years after the mission and documented thoroughly for a popular audience here for the first time.

In 1981, ground-based surveillance of satellites in orbit was rudimentary. Two Department of Defense facilities, in Hawaii and Florida, normally used to image Soviet and Chinese satellites, were now tasked to try to image Columbia in orbit. This was a daunting task: the shuttle was in a low orbit, which meant waiting until an orbital pass would cause it to pass above one of the telescopes. It would be moving rapidly so there would be only seconds to lock on and track the target. The shuttle would have to be oriented so its belly was aimed toward the telescope. Complicating the problem, the belly tiles were black, so there was little contrast against the black of space. And finally, the weather had to cooperate: without a perfectly clear sky, there was no hope of obtaining a usable image. Several attempts were made, all unsuccessful.

But there were even deeper black assets. The National Reconnaissance Office (whose very existence was a secret at the time) had begun to operate the KH-11 KENNEN digital imaging satellites in the 1970s. Unlike earlier spysats, which exposed film and returned it to the Earth for processing and interpretation, the KH-11 had a digital camera and the ability to transmit imagery to ground stations shortly after it was captured. There were few things so secret in 1981 as the existence and capabilities of the KH-11. Among the people briefed in on this above top secret program were the NASA astronauts who had previously been assigned to the Manned Orbiting Laboratory program which was, in fact, a manned reconnaissance satellite with capabilities comparable to the KH-11.

Dancing around classification, compartmentalisation, bureaucratic silos, need to know, and other barriers, people who understood what was at stake made it happen. The flight plan was rewritten so that Columbia was pointed in the right direction at the right time, the KH-11 was programmed for the extraordinarily difficult task of taking a photo of one satellite from another, when their closing velocities are kilometres per second, relaying the imagery to the ground and getting it to the NASA people who needed it without the months of security clearance that would normally entail. The shuttle was a key national security asset. It was to launch all reconnaissance satellites in the future. Reagan was in the White House. They made it happen. When the time came for Columbia to come home, the very few people who mattered in NASA knew that, however many other things they had to worry about, the tiles on the belly were not among them.

(How different it was in 2003 when the same Columbia suffered a strike on its left wing from foam shed from the external fuel tank. A thoroughly feckless and bureaucratised NASA rejected requests to ask for reconnaissance satellite imagery which, with two decades of technological improvement, would have almost certainly revealed the damage to the leading edge which doomed the orbiter and crew. Their reason: “We can't do anything about it anyway.” This is incorrect. For a fictional account of a rescue, based upon the report [PDF, scroll to page 173] of the Columbia Accident Investigation Board, see Launch on Need [February 2012].)

This is a masterful telling of a gripping story by one of the most accomplished of aerospace journalists. Rowan White is the author of Vulcan 607 (May 2010), the definitive account of the Royal Air Force raid on the airport in the Falkland Islands in 1982. Incorporating extensive interviews with people who were there, then, and sources which were classified until long after the completion of the mission, this is a detailed account of one of the most consequential and least appreciated missions in U.S. manned space history which reads like a techno-thriller.

September 2016 Permalink