Why NASA Can't Get It Right

By Jeffrey Kluger

Bread and salt were waiting for commander Eileen Collins and her crew when they docked the space shuttle Discovery with the International Space Station last Thursday. Station commander Sergei Krikalev had prepared the little ceremony, a Russian tradition intended to bring good luck to a visitor to your home. After the shuttle's stunning lift-off two days earlier--the first since the loss of the Columbia orbiter in 2003--it didn't seem the crew would need such happy charms. But now it appears the shuttle program as a whole--if not the astronauts themselves--may need a lot of luck indeed.

Only hours before the docking, NASA announced that the entire shuttle fleet was being grounded again, after evidence that four pieces of insulating foam--the largest the size of a skateboard--had spun off the ship's external fuel tank during lift-off, just the kind of debris that damaged Columbia's wing and doomed the ship. Only one small piece may have struck the shuttle this time, glancing off a wing with so little force it didn't register on impact sensors. But a camera mounted on the shuttle's 50-ft. arm as well as photos taken from the station have detected at least 25 dings in Discovery's insulating tiles. The most worrisome: a 1.5-in. divot near the nose, where temperatures can reach 3000DEGF.

So far, NASA insists that none of the chips present a danger, but in the days leading up to Discovery's planned Aug. 7 return, engineers--to say nothing of the astronauts' families--will be anxiously watching for any change in that assessment. If the shuttle were deemed unsafe to fly, the crew could take refuge in the space station, waiting out the five weeks it would take to roll out and gas up sister ship Atlantis for a rescue mission. But Wayne Hale, deputy shuttle manager, dismisses that scenario as "remote."

The larger question may not be the fate of Discovery but that of the whole fleet. After 2 1/2 years and $1 billion spent on safety upgrades designed to prevent just such a setback, how could things go so wrong again? Can a program that has already claimed the lives of 14 astronauts ever be safe? And what does any of that say about the Bush Administration's dream of sending crews to the moon and Mars?

No one knows why the foam flaked off in such large chunks. Since 2003, engineers have improved the way it's applied and contoured, and they eliminated it altogether in some tricky spots. That was supposed to ensure that no debris heavier than 0.03 lb. broke free. Cameras aboard Discovery suggest that the biggest chunk this time weighed just under a pound--nearly 30 times the limit. "The cameras worked well," said NASA chief Michael Griffin. "The foam did not."

Most of the debris came from a nettlesome area of the tank known as the protuberance air load (PAL) ramp, a ridge designed to minimize turbulence around cables and fuel lines. Tests and earlier flights convinced NASA that the PAL could withstand lift-off. "Obviously," says flight operations manager John Shannon, "we were wrong."

The chipped thermal tiles could be an even more serious problem. It's not certain whether foam caused any of that damage, but it's bad news all the same, particularly the 1.5-in. ding that is located on the edge of the door covering the nosewheel well. Any breach there could cause superheated gases to stream into the enclosed space where the landing gear is stowed, forming a vapor bomb inside the ship. NASA, however, is confident that all the chips are minor and is saying so with uncommon certainty. "It looks extremely good," says Shannon. "We don't have anything to worry about." Says Griffin: "The orbiter is a clean bird."

But it's not clear how NASA is going to fix the other birds in the shuttle flock. The tiles on Discovery seem relatively clean, with 80% fewer chips than shuttles usually sustain--a tribute to the strides the engineers have made. But had the big piece of foam broken off lower in the atmosphere, it would have slammed into a wall of thicker air and could have crashed far harder into the ship. NASA doesn't want to rely on luck again.

At a minimum, the PAL ridge will have to be redesigned. Currently, foam in that area is applied by hand instead of machine, which had seemed like the better method but clearly may not be. Engineers might also replace the ridge with a metal guard. "There is a very good chance the impact on the next flight could be minimal," says Phil Engelauf of the mission operations team. Still, the September flight of Atlantis will almost surely be delayed. How badly the schedule will back up after that is unknown.

Should NASA even bother with these finicky ships? It costs half a billion dollars every time a shuttle flies, and one major purpose this time was to transport 15 tons of supplies to the space station and off-load 13 tons of trash. A half-billion-dollar rubbish run is not, critics say, the best way to spend scarce space resources. What's more, given the ships' limited range, they won't play a part in future missions to the moon or Mars. Indeed, as long as the shuttle and station continue to consume about 40% of NASA's budget, those ambitious missions may never get off the ground.

For that reason, Griffin wants to decommission the fleet in 2010, meeting the U.S.'s commitment to its international partners to complete the space station, and nothing more. But the shuttle also means jobs--many in politically pivotal Florida--and that gives it a lot of support. In July, House Democrats negotiated a provision out of a budget bill that would have codified the 2010 deadline. And Science Committee chairman Representative Sherwood Boehlert of New York remains a loyal shuttle supporter. "Nothing is in jeopardy except the schedule," he said.

For now, it seems the shuttle program will plod on. NASA has always stressed that all spacecraft--even the shuttles--are unproven machines. "It's different from a commercial plane," says Hale. "[This is] an experimental vehicle." After 24 years, two lost ships and another scare this week, it may be time to ask if it's still an experiment worth conducting. --Reported by Stefano Coledan/ Cape Canaveral, Deborah Fowler/Houston and Eric Roston/Washington

ANATOMY OF A FLAW NASA spent more than two years and $1 billion to make sure that the accident that doomed Columbia in 2003 couldn't happen again. Despite assurances that it was safe to launch Discovery, several pieces of foam broke free during takeoff last week, forcing an embarrassed NASA to put all future flights on hold

WHAT WENT WRONG . . . The largest piece of foam missed Discovery entirely. Another piece may have struck the wing but did no obvious damage. Protective tiles on the orbiter's underside, including one near its vulnerable nosewheel well, seem to have been gouged during launch, perhaps by more falling foam

o UNDER THE SHUTTLE The orbiter didn't escape unharmed. There are at least 25 dings on its underside, including one near the nose, where temperatures can climb dangerously high o FOAM DEBRIS The biggest chunk was 2 to 3 ft. wide (0.6 to 0.9 m), about 1 ft. long (0.3 m) and up to 8 in. (20 cm) thick; it probably weighed 0.9 lbs. (0.4 kg), about half the size of the piece that downed Columbia o THE TROUBLESOME RAMP Most of the falling bits came from the so-called PAL ramp, a ridge of hand-sprayed foam designed to protect fuel lines from buffeting on takeoff o EXTERNAL FUEL TANK The tank, 154 ft. (47 m) tall and 27.6 ft. (8.4 m) wide, carries 535,000 gal. (2 million liters) of frigid liquid oxygen and hydrogen fuel. It is covered by polyurethane-like foam that keeps the fuel cool, prevents ice from forming on the exterior and protects the tank from overheating during launch

o FOAM -- 1 in. (2.5 cm) o Fuel o Aluminum casing o Epoxy

o Jettisoned external booster

. . .WHAT MIGHT BE DONE ABOUT IT . . . The Discovery crew will be testing several new ways NASA has devised to diagnose and fix launch damage before a shuttle returns to Earth o THE SPACE STATION Astronauts aboard the space station are taking high-resolution photographs of each shuttle before it docks with the station. That's one way the small "divots," or scrapes--including the one near the nosewheel well--were found on Discovery's underbelly o THE ROBOTIC ARM Discovery is equipped with a new 50-ft. (15.2-m) robotic arm that will reach out from the cargo bay to take infrared pictures of the orbiter at very close range, looking for potential problem areas o THE PATCH KIT Discovery's astronauts will test a tile-repair technique using a sort of paste--but not, unless absolutely necessary, on damaged tiles

. . . AND WHAT DANGERS LIE AHEAD During re-entry, Discovery's nose and the leading edges of the shuttle's wings--and the atmosphere surrounding them--could heat to as much as 3000-oF (1650-oC). Any chink in the shuttle's heat-resistant armor could allow superheated air to penetrate unprotected areas, where it would act like a blowtorch--the same process that destroyed Columbia o TILE DAMAGE One gouge is close to the wheel well, where an influx of hot gases could cause an explosion

With reporting by Stefano Coledan/Cape Canaveral, DEBORAH FOWLER/HOUSTON, Eric Roston/Washington