Independent aerospace safetyexperts, meanwhile, are applauding the three-week postponement a delaythat comes amid an intense agency push to ramp up NASAs $60 billion InternationalSpace Station construction project.

"It was a very sensible,safe thing to do," said Seymour Himmel, a member of the Aerospace SafetyAdvisory Panel, an independent group charted by Congress to oversee NASAaeronautical and spaceflight programs.

"Its not an easy thing with all the flights lined up for station construction to roll the shuttleback to its assembly building and take a real [schedule] hit. But I thinkNASA did the right thing."

Whats more, Himmel and otherexperts say the slip to Feb. 6 is representative of a conservative agencymindset that is significantly different than the one which led to the January1986 Challenger explosion, which killed seven astronauts.

"Its a far cry from violatingyour own [safety] rules, which was done when they agreed to fly Challenger,"Himmel said.

Perched atop a giant trackedtransporter, Atlantis crept off its launch pad and into KSCs 52-storyVehicle Assembly Building so that NASA can inspect suspect cabling associatedwith its solid-fuel rocket booster-separation system.

The cable concerns aroseduring exhaustive inventory inspections that were ordered after a seriousglitch cropped up during the Nov. 30 launch of sister ship Endeavour ona station construction mission.

On that flight, a pyrotechniccartridge designed to separate the shuttles left-hand solid rocket boosterfrom its 15-story external tank failed to fire. A backup cartridge workedas advertised, and the booster separated cleanly.

Had the backup failed, however,Endeavour and its five-man astronaut crew almost certainly would have beenkilled in a Challenger-like catastrophe.

The reason: A strut attachingthe booster to the lower part of the tank would have failed to separate.The massive fuel reservoir would have ruptured as a result, triggeringan explosion a little more than two minutes into flight.

Next page: The resultsof 'wiggle tests' and other inspections

~

Engineers subsequently tracedthe problem to a faulty cable designed to route critical computer commandsto the pyrotechnic cartridge. Clear evidence of damage due to normal wearand tear had gone undetected during preflight inspections.

The close call prompted extrainspections on separation system cables hooked up to Atlantis twin boosterrockets, and the shuttle ultimately was cleared for a Jan. 3 move to launchpad 39A.

At the same time, though,exhaustive inspections of all separation system cables in NASAs inventorywere being conducted. Four faulty cables subsequently were discovered duringthose inspections, raising new concerns about the integrity of Atlantisseparation system.

Like the cables on Atlantis,the suspect quartet from stock inventory had passed X-ray examinations.The inventory cables, however, failed extra tests designed to simulatethe ability to relay crucial separation commands when the wiring is subjectedto launch-like vibrations.

The cause of the test failuresstill is not understood.

The Atlantis cables, meanwhile,had not been put through the extra "wiggle tests," which call for the wiresto be jerked from side to side and up and down while electrical power isbeing transmitted through them.

And that fact raised redflags with many within NASAs shuttle program.

"We could not get comfortablewith the fact that what we had at the pad had not been [put through wiggletests]," NASA shuttle program manager Ron Dittemore said.

Thats not to say, however,that there wasnt a spirited discussion over the matter.

With the Atlantis crew waitingto fly to KSC and countdown clocks poised to pick up, a teleconferencewas convened Monday with officials from KSC, Johnson Space Center in Houstonand Marshall Space Flight Center in Huntsville, Alabama.

Engineers presented testdata that showed 194 cables containing 3,028 electrical conductors hadbeen examined. Problems, however, only were discovered with four of theconductors, which are designed to relay separation commands.

Whats more, three of thosefour conductors were spares that only would have been pressed into serviceif primary conductors failed to route signals to pyrotechnic charges.

Some engineers, consequently,felt the odds of a failure in flight were remote at best, and that NASAshould press ahead with its launch plans.

"We had some statisticalanalysis that we had performed that indicated, based on statistics alone,that you might be comfortable proceeding with flight," Dittemore said."And there were a number of folks that felt comfortable with just a statisticalanalysis."

But engineers from Marshall where the booster project is managed argued for extra inspections tomake certain the cabling on Atlantis isnt prone to failure.

"There was another camp thatfelt that even though the statistical analysis [showed] we might be justifiedto proceed, the hardware was telling us something," Dittemore said.

"We had four failures herethat we discovered in our testing, we dont understand yet why, and [theyfelt] we ought to listen to what the hardware may be trying to tell usand do further inspections."

In the end, Dittemore said,"the prudent thing to do was to go back and inspect."

The decision-making process-- and the ultimate decision -- stand in stark contrast to the fatefulcall made by NASA managers the night before Challengers 10th and finalflight.

Faced with an equally ambitiousshuttle flight schedule, engineers from NASA Marshall and booster builderMorton Thiokol Inc. held a teleconference to discuss the ramificationsof record cold weather on rubber O-ring seals in Challengers solid-fueledrockets.

Thiokol engineers were concernedthat the O-rings which are designed to prevent hot gases or flame fromescaping the boosters might not work properly.

Postflight inspections after12 of 24 shuttle missions to date had uncovered clear evidence of significantO-ring erosion, and the sensitive seals were not certified to fly in weatherbelow 53 degrees Fahrenheit (11.5 degrees Celsius).

While the Challenger crewslept, Thiokol engineers unanimously recommended a launch postponementbecause they feared the O-rings might allow hot gases to leak in temperaturesthat ultimately dipped to 36 degrees Fahrenheit (2.2 degrees Celsius) atlaunch time.

Mid-level managers from Marshall,however, protested strongly, saying Thiokol engineers in spite of thepast history of O-ring problems -- did not have enough data to supporttheir concerns.

Bowing to pressure, Thiokolengineers eventually reversed their recommendation and approved plans toproceed with launch. Higher-level NASA managers, meanwhile, were not informedof the late-night debate and the Challenger crew met its fate.

The presidential commissionthat investigated the accident found that there was a "serious flaw" inthe decision-making process leading up to the Challenger launch.

NASA failed to flag risingconcerns about O-ring problems, waived its own strict launch rules andallowed schedule pressure to override flight safety issues, the commissionfound.

Things are different now,especially when technical issues crop up with so-called "Criticality 1"systems those systems whose failure would lead to the loss of a shuttleand its astronaut crew.

"Now the whole discussionrevolves around the technical merit [of an issue], the pros and cons interms of flight safety, and then a decision is made," said former NASAlaunch director Bob Sieck, who also is a member of the Aerospace SafetyAdvisory Panel. "Schedule ramifications are kind of an after-the-fact thing."

The decision to delay theAtlantis flight is just one case in point. Another was the decision toground the agencys shuttle fleet when dangerous orbiter wiring flaws wereuncovered in 1999.

"The sensitivity to safetyis as high as Ive ever seen it," said Sieck, who joined NASA in 1963 andspent more than a quarter-century with the agency. "People are very sensitiveto the criticality of what they do and what they work on, and they performaccordingly."

That, Dittemore said, isjust the way NASA is going to do business in a day and age when the agencycan ill-afford another Challenger-type disaster.

"I guard against this phenomenonof 'go-fever like it was the plague," Dittemore said.

"Schedule is schedule andit does not compete with safety. We will roll back from the pad as wedid in this case and we will take whatever other steps are necessaryto preserve the integrity of the [shuttles] and the safety of both thosewho fly the vehicle and process it on the ground," the shuttle programmanager added.