The space machine that carried teacher Christa McAuliffe and six other astronauts to their deaths Jan. 28 was a classic aerospace compromise balancing the abilities of technology with the realities of expense.
That compromise produced the remarkable shuttle, which carried astronauts safely into space 24 times to launch and repair satellites and undertake hundreds of scientific experiments. But it also built a pattern of success that dulled the senses to extraordinary danger and led to catastrophe on the 25th launch.
Like many of mankind's most stunning failures, the Challenger disaster was engendered years earlier in a complex sequence of events and decisions, some of which became unwitting but lethal mistakes.
This article, the first in a periodic series on the National Aeronautics and Space Administration and the American drive into the heavens, will examine the process of designing the shuttle and how NASA arrived at a machine that resembles a DC9 strapped to a giant thermos bottle and two Roman candles.
"In our business, we consider several factors," said Maxime A. Faget, a key figure in the design of the shuttle. "We design for performance, we design to minimize cost, we design to keep weight down, and finally, you have to make it safe. It is obvious that if you put all the safety features you would like, you would probably never get to orbit."
Now, 14 years and $31.3 billion after President Richard M. Nixon approved the space shuttle program, the Challenger has exploded and with it NASA's reputation for skill, care and safety.
In fact, its highly burnished image never conveyed reality, according to many present and former NASA officials, aerospace industry executives and other NASA watchers. Instead of being a well-oiled machine, all for one, NASA since its creation in 1958 has been a collection of diverse interests and competing power centers that have managed most of the time to overcome their differences and do marvelous things.
Nonetheless, it was considered a given that "safety first" was more than a slogan. Even that belief has been shattered. As recently as Thursday, a top space agency official told the Rogers Commission investigating the disaster that the shuttle program was beset with major management flaws.
The National Space Transportation System, as the shuttle is officially named, was conceived by a committee in part to meet the growing need for launching military and commercial satellites, supposedly at reduced cost. There also were less explicit reasons, such as keeping a manned space effort alive and retaining the precious skills that thousands of NASA and contractor employes had developed.
Their successes with flights of mythological designation -- Mercury, Gemini and Apollo -- had carried mere mortals named Shepherd, Glenn and Armstrong to the shores of the rest of the universe, built for NASA a wondrous reputation and placed into the vernacular the notion that if man can fly to the moon, anything is possible.
NASA has had disaster before, notably the 1967 fire in the Apollo One crew capsule that killed three astronauts during training. A reading of the accident report 19 years later shows that the fire had some of the elements that investigators are finding in the shuttle explosion: marginal workmanship, inattention to detail and an assumption that something that worked reliably -- an escape hatch -- also would work expeditiously, which it did not.
That investigation was an in-house job, with its key determinations softened by bureaucratese. For example, "The spacecraft pad work team was not properly trained or equipped to effect an efficient rescue operation under the conditions resulting from the fire."
The investigation of the Challenger disaster, however, is being led by a prosecutor, former secretary of state William P. Rogers, whose team is pointing fingers and naming names. There is surprise inside and outside the agency at the depth of the problems that took safety out of the cockpit and turned it into a passenger.
Blue-Collar Name and Workhorse Profile
On Jan. 5, 1972, with the moon conquered and only two Apollo flights remaining, Nixon announced that the United States would build "an entirely new type of space transportation system," one that would revolutionize travel into space "by routinizing it."
"It will take the astronomical costs out of astronautics," Nixon promised.
Instead of invoking the gods of mythology, this space machine was called by a blue-collar name, the shuttle. It was based on a concept at least 30 years old and in typical Washington fashion was to be all things to all people: launch platform for military and communications satellites, laboratory for science and industry and a money-saver. Not only that, it would solve the problem of how to build and resupply a long-desired outpost, the space station now on the drawing boards.
The shuttle's major feature, one that gave it great advantage over the thundering one-shot rockets that lifted Apollo, was that it could be used again and again, making it significantly cheaper. It is debatable whether that has been achieved and the Office of Management and Budget estimates the shuttle's cost overrun at about 30 percent after inflation.
Three NASA power centers share responsibility for the shuttle. They are the Johnson Space Center in Houston, which trains the astronauts and supposedly has overall control of the shuttle program and the mission; the Marshall Flight Center in Huntsville, Ala., which is in charge of the rocket engines (liquid fuel) and rocket motors (solid fuel) that propel the shuttle into low-earth orbit, and the Kennedy Space Center in Cape Canaveral, Fla., which assembles and launches the shuttle. The instant the shuttle clears the launch tower, control of the mission switches from Kennedy to Johnson.
All centers are equal, but Johnson is more equal, according to many NASA veterans and retirees. Johnson gets most of the television time, and the astronauts radio their reports to "Houston."
From the earliest days of manned spaceflight, "The Marshall guys were not what I would call cooperative with the Johnson guys," a former NASA senior official said. "Culturally, the rift started there, there was no love lost." Marshall once was first, he said, because of Werner von Braun and his German scientists and their rocket work, "But by the time Apollo was over, Marshall was in second place."
Weight Concerns and Mystery Payload
Prime responsibility for developing the shuttle rested with Johnson, although Marshall had the job of finding the power necessary to boost 65,000 pounds of payload into an orbit over the equator, or 32,000 pounds into an orbit over the poles.
A shuttle launched from Cape Canaveral into an equatorial orbit takes advantage of the Earth's rotation to help kick it into space. A shuttle launched from California's Vandenburg Air Force Base into polar orbit would have no such advantage and thus must pay the price -- a smaller payload.
Weight is the issue in manned flight, whether it is hang gliders or spacecraft. Anything that can be done to reduce weight of the vehicle increases the weight that can be carried for profit. As the shuttle's designers set to their task in the late 1960s, they were attempting to balance intricate requirements, including payload weight, reusability, cost and safety.
When they were through with their work by the end of 1971, they thought they had a vehicle that would meet those needs, though they knew they were involved in a risky business, using highly explosive fuels and rockets that become uncontrollable once fired. They also knew that, no matter how hard they tried, they would not anticipate every possible problem.
There were many early concepts for the shuttle, but the serious study began in 1969 with final decisions reached in 1971, shortly before Nixon's announcement in January 1972.
The designers' first challenge was to make sure they had the Pentagon as a customer to help underwrite their shuttle, and the Air Force wanted a big payload deliverable from a large cargo bay.
"In general it was felt the bigger the volume and weight the better it would be," Faget said in an interview. He is now president and chief executive of Space Industries Inc. in Houston but was the director of engineering and development at the Johnson Space Center from 1962 to 1981.
"Extensive studies were made, and we made the payload bay about as big as we thought we could," Faget said. "The Air Force was very interested in not making it any smaller, that seemed to be their point of pain."
Although one of the prime designers, Faget said he never learned what the Air Force wanted to carry. "They alluded to some missions they had, something that needed that much room. This was at a very high level," Faget said. The Air Force referred all questions to NASA, which said it did not know, either.
To carry the weight needed led to a choice between a cargo bay about 14 feet by 45 feet or 15 feet by 60 feet. The latter was chosen because it worked better when fitted into a flying machine, according to NASA's Joseph P. Loftus. That's enough room to carry three or four communications satellites.
Since the goal was to build a fully reusable device, the shuttle had to have rocket engines to escape the Earth's gravity and atmosphere, and wings to fly when it came home again. The designers had determined through earlier space missions that the only way they were going to get up was by having more than one rocket stage.
Because the shuttle requires tremendous thrust, it needs an enormous fuel load, leaving little room for anything else. Fuel approaches 90 percent of the weight of the shuttle, which is 4.5 million pounds at liftoff.
"If you want to get to orbit, it takes 30,000 feet per second of acceleration," Faget said. "You encounter drag from the atmosphere and have to overcome the force of gravity. So a rocket is very inefficient when you first lift off, it is just overcoming gravity.
"If you stage, you only have to have 15,000 feet per second" in each stage, he said. That reduces the amount of propellent needed in the upper stage, because it does not have to carry the weight of a lower stage after it is jettisoned and, by the time that occurs, the atmosphere is much thinner and provides much less resistance.
So NASA's first ideas showed an orbiter, looking much like the one we know today, resting atop a much larger beast that looked almost identical. The idea was for a piloted first stage to get the shuttle going, separate from the second stage and return in an airplane-like landing. The second stage, containing the payload and the orbiting crew, would complete the mission, then return to Earth.
Both stages would be driven by engines fueled with liquid hydrogen and liquid oxygen, just as the shuttle's main engines are today, and both stages would carry their own fuel tanks internally.
But the designers soon encountered problems. For one thing, the prototype got heavier and heavier, tipping the scales at more than 5 million pounds. There were also worries about carrying the propellent internally.
"There were all kinds of heat leaks around," Faget said. "If you have a combustible hydrogen mixture in the air, I've been brought up to say you'll have an explosion. In a lot of hydrogen explosions they have never found an ignition source, but they have always found a critical mass of hydrogen."
The reusable, fly-back first stage was scrubbed.
That led to another concept. The orbiter's fuel tanks would be placed outside, one on each side of the orbiter, and the whole business would be launched on the back of a pilotless first-stage booster, never further defined. The computers didn't like the aerodynamic pressures and loads such a configuration would create, so it too was abandoned, but the idea of external fuel tanks was solidly in place.
Then the designers dreamed up an enormous first-stage liquid booster to carry both the orbiter and the necessary fuel. The booster would have internal hydrogen and oxygen tanks and, to meet the goal of reusability, would be parachuted back to Earth for recovery.
But the giant booster would have weighed more than 800,000 pounds, and it was questionable whether it could have been parachuted back for recovery. Another drawback was that the orbiter's engines were not to be fired until the mission was well under way; if there was a start-up problem, there was no way to catch it before launch.
"Then we went to what we got," Faget said. "We could light all the engines on the pad, a big advantage. That gives you a complete readiness check and the ability to shut down if you need to."
The weight of the tanks needed for the second stage would not have to be carried into orbit because the external tank would be jettisoned. The two spent solid-rocket boosters would weigh about 200,000 pounds each on a parachute, making them recoverable.
Solid boosters were chosen because solids provided the best way to get maximum thrust for minimal weight, according to several rocket specialists. A liquid rocket for the same task would have been much heavier and larger.
O-rings, washer-like seals between the segments of the solid rocket boosters that are the prime suspect in the Challenger disaster, were never considered a major problem, according to the designers. O-rings had been used for years on rockets to seal all kinds of joints, and the highly successful Titan rocket was segmented and had worked reliably with just one O-ring, not two as the shuttle rockets carried. An attitude was firmly in place 15 years ago, according to NASA veterans, that the solid rockets were not a major worry.
Eliminations and The Final Product
Several safety features in the orbiter's design were dropped and have become the subjects of second-guessing after the Challenger explosion. They include escape rockets to carry the orbiter away from the boosters and external tank if a problem developed early in the launch, ejection seats for the crew and air-breathing engines to provide a powered landing when the orbiter returns instead of the glider-type landing it makes now.
Escape rockets and auxiliary engines "were dropped as being not effective," Faget said. "Weight growth was not the driving issue in either one of these cases."
Ejection seats were used on early shuttle flights but they, too, were dropped. "For vertical flight, ejection seats are very limited," said NASA's Loftus. "When at cruise altitude," miles high, "it was impossible to have an ejection system." A classic aviation tradeoff: Something nice to have, but of limited use. If removed, obviously, weight is reduced.
There were many other decisions, including whether to use a conventional wing or a delta wing, which permits supersonic performance and much longer glide-range from orbit and was required by the Air Force; whether to use heat-resistant tiles or shingles to protect the orbiter from the heat of reentry; whether the rockets should be fueled by liquids or solids or both.
After all this pencil and computer work, the shuttle we know today resulted. Its major components are:
*A winged orbiter including a cockpit, payload bay and three primary liquid-fuel rocket engines, about the size of a twin-engine passenger jet, the McDonnell Douglas DC9.
*An enormous external fuel tank, as tall as half a football field is long. It looks like a thermos bottle for good reason, because it is heavily insulated to help contain the liquid hydrogen and liquid oxygen that are mixed to fuel the orbiter's engines. The external tank is also the structural backbone of the shuttle at launch, holding both the orbiter and two solid fuel booster rockets.
*Two boosters, each about as tall as the external tank, strapped to the sides of the tank to provide extra thrust at liftoff.
On the launch pad the whole vehicle rests on the bases of the two solid rocket boosters. At launch, the three main liquid-fueled engines are fired first and carefully checked by computers; if the computers don't like what they sense, they turn the engines off and abort the flight, which has happened once.
If the computers say A-OK, the solid rockets are fired, the point of no return. The beast thunders off the pad, reaching the speed of 3,094 miles per hour in two minutes.
On normal flights, after two minutes the shuttle has reached an altitude of 28 miles, where the solid rockets burn out and are separated from the orbiter. They fall back to Earth on parachutes and are retrieved.
The three main engines continue firing for about eight minutes until the shuttle reaches an altitude of 68 miles and a speed of 17,440 miles per hour. Then the external tank is jettisoned and falls, unclaimed, into the ocean. Smaller liquid-fuel engines on the orbiter complete the job of placing it in orbit.
The Challenger flight 10 weeks ago ended in a roar 73 seconds after liftoff, apparently when the now-notorious O-rings in the right-side solid rocket boosters allowed hot gases to burn through the casing, creating, in effect a rocket firing in two directions, sideways and up. It didn't take long for the bottom bracket holding the booster to the external tank to snap and for the booster's nose to pivot into the external tank and produce the inevitable explosion in the presence of hydrogen.
"It was never discussed, but we knew if a solid rocket failed it would be catastrophic," Faget said.