<

At the dawn of the U.S. shuttle program, NASA engineers weighed the risks posed by nagging failures in the twin rocket boosters that propel the spacecraft into orbit.

They quickly discovered severe heat damage in the O-rings designed to stop searing gas spewing from joints in the rockets. NASA weighed the risks and decided the flaw was ''tolerable,'' an analysis that held for 24 flights.

Then a plume blew through the O-ring seals on Challenger Jan. 28, 1986, setting off an explosion that killed seven astronauts and grounded NASA for three years.

Seventeen years later, the investigation into the Columbia disaster may be revealing a similar pattern of miscalculations and misjudgments about one of the shuttle system's most persistent technical problems -- shards of insulating foam damaging the shuttle's vital and vulnerable thermal armor.

''The same kind of mentality seems to be at work,'' said Diane Vaughan, a Boston College sociology professor who wrote the 1996 book The Challenger Launch Decision. ``You go from not expecting any damage to expecting some, to a point where you´re seeing damage all the time.

``Rather than see it as a warning sign, you see it as an acceptable risk.´´

NASA never acted on a string of internal studies calling for improved testing to reduce foam ''de-bonding'' from the giant aluminum external fuel tank that powers the shuttle.

The foam insulation around the tank kept flying off, sometimes damaging lifesaving thermal tiles that keep the shuttle from burning up. Investigators are examining whether that happened with Columbia, with fatal consequences.

One 1995 report estimated 90 percent of damage to thermal tiles on the shuttle's belly came from foam debris alone.

''Ninety percent? That's scary,'' said Charles McMahon, a professor of materials science and engineering at the University of Pennsylvania. ``It´s scary if any serious damage is coming from that.´´

NASA officials say it's premature, and faulty, to compare the Columbia to the Challenger. ''It's a large leap at this point in the investigation,'' spokesman Bob Jacobs said.

INDEPENDENT PANEL

An independent panel investigating the Columbia accident on Thursday disclosed that super-heated air likely seeped through a breach in Columbia's left wing or wheel well, triggering the shuttle's ultimate collapse.

The panel could not say whether the breach was caused by a structural tear in the shuttle's aluminum frame or a hole caused by debris -- either the foam or space junk.

NASA engineers and the independent panel investigating the Columbia accident will run through a range of possible causes, but that 2.67-pound chunk of foam that dislodged from the tank and struck the left wing 81 seconds into flight remains an obvious suspect.

Even if it turns out not to be the root cause, the accident has already forced NASA managers to reassess the blanket dismissal of insulation as a potential hazard.

It also raises the same sort of questions about NASA's handling of safety issues heard in 1986. They echo especially loudly for Lawrence Mulloy, a former NASA engineer who made many of the key calls on Challenger's O-rings and was sharply criticized by a presidential commission investigating the catastrophe.

If -- ''a big if,'' he stresses -- foam debris proves the trigger for Columbia's breakup minutes from landing in Florida, then he fears NASA forgot the lessons of its own history.

`ACCEPTABLE´

''If it turns out that's the case, you have a direct comparison to the acceptance of the O-rings,'' he said. As early as the second shuttle flight, he said, engineers found heavy damage on those critical seals.

''We saw the tests and analysis and convinced ourselves that the O-ring was doing the job,'' said Mulloy, who retired not long after the Challenger disaster and now lives outside Nashville. ``We convinced ourselves that this was an acceptable situation and it did not constitute a safety-of-flight risk.´´

SAFE FLIGHTS

Mulloy hears the same sort of rationalization he once employed coming again from NASA. Foam debris, though a problem from the very first tests of the shuttle, became less of a concern as more flights returned safely, despite often heavy damage to the thermal tiles. After 113 flights, he said, foam would have moved far down the list of concerns.

''You build up a history and a database that says you can tolerate the deviance you're seeing again and again,'' he said. ``That was the mistake on Challenger. I hope it was not the mistake on Columbia.´´

Mulloy, chief engineer for external tanks from 1973 to the first Columbia launch in 1981, says foam loss was always a worry. The external tank lost foam on the very first shuttle mission in 1981 and, as recently as the October launch of Atlantis, a large patch struck one of the rocket boosters. Atlantis landed without problems or significant damage.

Most other flights also have experienced foam loss during liftoff, some suffering hundreds of hits to thermal tiles. NASA routinely dispatched crews to hunt for bits of foam on Florida beaches following launches.

Experts say the foam can fail for a number of reasons. Thermal cracks -- fissures that occur as the metal tank constricts from the super-cooled fuels -- can allow ice to form and peel the foam from the tank. The massive vibration, heat and aerodynamic stress of the first few minutes of liftoff is a major concern. One 1983 NASA report examined the problem of ''spallation'' -- chunks falling away -- from pressure buildup from heating during ascent.

CONTRIBUTING ERRORS

Installation errors can contribute: Dirt, oil or even ''burps'' in spray guns can cause pockets of loose material. Close-out areas -- small sections sprayed on by technicians at the Kennedy Space Center launchpad -- were a particular problem.

Despite numerous changes in foam formulas and application techniques by contractors, foam falloff has never been fully controlled.

One 1995 NASA report recommended using shearography to reduce damage to the orbiter belly from foam.

The technique employs a laser and vibration or pressure to detect cracks and weak bonds under the foam with a degree or accuracy approaching 95 percent, said John Newman, whose company, Laser Technology Inc. of Norristown, Pa., consulted with NASA on the problem since 1992.

ALARMED

As recently as last year, alarmed by repeated foam failures, an engineer at Kennedy Space Center wrote a report urging standard shearographic testing.

But NASA never moved beyond the testing stage.

NASA has had numerous other warnings of the potential threats of foam.

As far back as 1990, two researchers concluded that damage from foam debris was a leading source of potential thermal tile failure. In 1997, after an orbiter returned with more than 300 hits from debris, a senior NASA engineer wrote a report detailing significant damage to the tiles.

SPEED RELEVANT

''Foam [causing] damage to a ceramic tile?!'' wrote NASA engineer Greg Katnik. ``That seems unlikely. However, when that foam is combined with a flight velocity between speeds of Mach 2 to Mach 4, it becomes a projectile with incredible damage potential.´´

Although NASA Administrator Sean O'Keefe said Wednesday during congressional testimony that the foam in question was traveling at a rate of roughly 50 mph, his comments don't jibe with reports by NASA and Boeing engineers.

O'Keefe described the foam's impact as ''a Styrofoam cooler blowing off a pickup truck ahead of you on the highway.'' Yet a Jan. 21 study, now referred to as the Boeing study, put the speed of the foam between 400 and 500 mph.

Vaughan, the Boston College professor, said the Columbia disaster had revealed enough concern about foam that NASA will be forced to finally deal with it or develop a new technology.

''It's a culture in which flying with problems is normal and routine,'' she said. ``Where thinking becomes distorted is when it becomes so routine they no longer see it as a critical flight issue.´´