|
Nobody saw shuttle smoking gun { August 24 2003 } Original Source Link: (May no longer be active)
http://www.washingtonpost.com/wp-dyn/articles/A37325-2003Aug23.htmlhttp://www.washingtonpost.com/wp-dyn/articles/A37325-2003Aug23.html
Columbia's 'Smoking Gun' Was Obscured
NASA Did Not See a Deadly Risk When Foam Struck Shuttle Wing
By Kathy Sawyer
Washington Post Staff Writer
Sunday, August 24, 2003; Page A01
Nobody saw the "smoking gun" when it fired, 81 seconds after the space shuttle Columbia lifted off from its Cape Canaveral launch pad on Jan. 16.
The shuttle was over the Atlantic Ocean, racing skyward at more than 1,600 mph -- greater than twice the speed of sound -- as it climbed through an altitude of 66,000 feet in the thinning air. A ragged section of insulating foam -- a Styrofoam-like slab weighing almost two pounds and about the size of a flat-screen TV -- flew off the shuttle's 15-story-tall, bullet-shaped propellant tank.
The fragment hit the roiling supersonic air stream like a Styrofoam cup tossed from a racing car, slowed abruptly and started spinning rapidly as it traveled about 60 feet. In a split second, it struck the leading edge of the left wing at a relative velocity of 530 mph. Two smaller foam chunks that broke off at the same time missed the wing.
Even though it seemed insubstantial, the foam Frisbee, traveling at that velocity and spinning at least 18 times a second, struck with about a ton of force. It was simple physics. The impact blew a hole probably six to 10 inches across in the leading edge of the left wing.
While most of the ingredients of Columbia's destruction had been collecting for some time, tucked away in the crannies of a complicated bureaucracy, no one had seen them, either. In contrast to the 1986 loss of the shuttle Challenger, when engineers had tried to stop the launch only to be overruled by higher-ups, this time the portents had become invisible to all those who might have altered events.
The catastrophe that had begun over the Atlantic, dooming Columbia and its seven astronauts, would not unfold for 16 more days -- also mostly out of sight.
It would take armies of engineers and six months of arduous detective work to bring into focus what had gone wrong on Feb. 1, not only with the shuttle but with the organization that had sent it on its way.
The following is an account of what most likely happened to Columbia, based largely on the results of that effort -- official records, transcripts and video from NASA and the Columbia Accident Investigation Board, including Freedom of Information Act releases and engineering test results and analyses -- and interviews. The probe will culminate this week when the independent board releases its final report exposing in detail the roots of the tragedy.
Until the end, Columbia's mission would seem as close to perfect as the risky business of flying into space ever gets.
A Haunted Moment
As the main engines ignited, the astronauts heard a crack followed by a muffled roar and felt the vibrations as their 4.5-million-pound craft started skyward, riding a controlled explosion. It was 10:39 a.m. on Jan. 16.
The launch is the tensest part of any mission. Still, Cmdr. Rick Husband, 45, a colonel in the Air Force who was strapped into the front left seat, took note of the brilliant sunlight streaming in the overhead windows, the shadows moving over the control panels as the shuttle rolled onto its back.
"Columbia, Houston. You're go at throttle up," came the call from mission control in Houston, 70 seconds after liftoff.
"We copy, go at throttle up," responded Husband, a former test pilot who had flown an earlier shuttle mission. That phrase -- and that moment in the shuttle's fiery ascent -- had been haunted for 17 years. This was the point at which Challenger had exploded, etching America's first in-flight space disaster in sudden fire and smoke across another January sky, 88 shuttle missions earlier.
The commander's acknowledgement of a routine power adjustment came just after the spaceship had broken the sound barrier.
This time, another 10 seconds passed before the trouble started, unheeded. The errant foam projectile broke a hole in the crucial material known as RCC, for Reinforced Carbon Carbon, a tough material that was supposed to protect key surfaces from the lethal heating that occurred when the space plane reentered Earth's atmosphere at more than 24 times the speed of sound.
At the time the foam struck, there was still time for the shuttle to abort its ascent and make an emergency landing -- a risky move that no shuttle has ever attempted. But the shuttle, although intently monitored during its ascent, had no instruments to immediately detect the fatal blow.
Another day would pass before engineers discovered the impact in a blurry video of the launch. And even then, no one would appreciate its dire implications.
As the shuttle raced toward orbit, however, onboard sensors did record possible signs of the trouble. Unusual temperature and pressure readings registered in the wing beginning about two seconds after the impact. But these signs were ambiguous and, in any case, could not be assessed in "real time." They were being stored by a data recorder for playback after the flight. (More than a month later, on March 19, searchers would find the recorder and its wealth of stored evidence embedded in muddy ground near Hemphill, Tex.)
Eight minutes and 26 seconds after liftoff, Columbia reached orbit for the 28th time in its 22-year history. The main engines shut down and the orbiter ditched the drained tank. At that point, Air Force Lt. Col. Michael Anderson, 43, and Navy Capt. David Brown, 46, got out of their seats to take pictures of the event. Their film, stored on board for post-landing analysis, would never be found.
As she reached weightlessness, astronaut Laurel Clark, a Navy flight surgeon, 41, noticed "the zippers and all the belts that have D-rings . . . are always floating and hitting each other and jingling, and it makes this beautiful tinkling music in the background all the time."
A Long Road to Liftoff
Columbia's crew of seven had spent an unexpectedly long time -- three years -- training for this mission and this moment. Four of them were rookies -- Navy Cmdr. William "Willie" McCool, 41; Israeli Air Force Col. Ilan Ramon, 48; Clark and Brown. The other three -- Kalpana Chawla, an aerospace engineer and pilot, 41, Husband and Anderson had each made one previous shuttle flight.
Ramon's presence had heightened the media focus on security at the launch. The possibility of a terrorist act would spring into many people's minds 16 days later.
There had been times when it seemed this particular mission was destined never to leave the ground. The launch date had slipped eight times. Most recently, in the spring of 2002, shuttle officials had grounded the entire fleet to deal with potentially lethal cracks in fuel-line hardware -- one of many instances where the system had worked well.
The astronauts knew they would be working around the clock in two shifts to carry out one of the more complex shuttle missions of recent years. On board with them were 13 rats, eight garden orb weaver spiders, five silkworms, four Medaka fish eggs (and other fish types), three carpenter bees, 15 harvester ants and several canisters of pinhead-sized round worms. In April, the worms would be found alive in their containers, the flight's only survivors.
The critters were part of the more than 80 experiments packed aboard the orbiter.
Ordinarily, the shuttle would have carried its giant robotic arm mounted in the cargo bay, which could be used like a cherry picker to support an astronaut working outside the spaceship, but its place was taken by a special research facility. Its absence, and the fact that the shuttle was not visiting the orbiting international space station, eliminated the more practical ways to inspect for possible wing damage.
NASA had designed Columbia's mission under pressure from members of Congress and scientists who wanted research in weightlessness to continue while the space station is under construction. After the accident, critics would say none of the onboard research was worth lives. But the astronauts had accepted the risk.
Foam Impact in Videos
Early Friday, the morning after the launch, Linda Ham in Houston held the first meeting of her mission management team -- a group of senior shuttle managers responsible for key decisions and supervision of all aspects of the flight. It was the first of only five such sessions during the 16-day flight, even though the rules called for the group to meet daily. And those five meetings were actually teleconferences, with some participants linked in from scattered locations.
That day, the group discussed a few minor problems, such as the failure of a fuel cell tank heater. The foam impact never came up. Engineers would discover it that same day in a review of long-range tracking videos of the launch. The only view of the area where the foam hit showed no details smaller than two square feet and did not reveal whether damage had been done.
Later that day, about 24 hours after the launch and shortly after the orbiter had fired its maneuvering jets, roughly between 10:30 and 11:00 a.m. EST, an object floated slowly away from the orbiter.
Here was another clue that something was awry, but no one saw the object depart. The astronauts were changing shifts and talking with controllers on the ground. Not until a little over a week after Columbia was lost would an inspection of Air Force radar data reveal the mystery debris. In the months that followed, painstaking tests and analysis of the object's ballistic properties and radar signature, compared with various bits of shuttle hardware, would narrow the plausible list of what the object was to certain components of the wing's leading edge.
Finally, in July, in a key test designed to duplicate the launch-day impact, the foam projectile would blow a hole the size of a manhole cover in the wing front, leaving fragments of the RCC resting inside the cavity. The properties of several shards matched those of the mystery object.
No 'Safety of Flight' Issue
Over the long holiday weekend honoring Martin Luther King Jr.'s birthday, the mission managers suspended their meetings, and some participants took time off. They met next on Tuesday, Jan. 21, five days into the flight.
Investigators would later determine that, if Columbia's crew was to be saved before their life-support ran out, NASA officials would have had to know by this date that they faced a certain catastrophe. Only then would they have enough time to mount their best chance at a rescue -- a radical, risky emergency launch of another shuttle and crew.
It was at this session that the issue of the foam strike came up for the first time, at the end of a list of minor concerns.
Ham said, "All right, I know you guys are looking at the debris." She was referring to the photography that had captured the foam strike.
In some ways, it was a familiar problem. Shuttle engineers had gotten used to small "popcorn" pieces of tank foam coming loose and dinging the orbiter's heat-shielding on every mission. Efforts to stop it had failed.
Columbia's launch involved a rarer and more serious type of foam shedding, from an odd-shaped fixture called the bipod ramp, where struts attach the tank to the bottom of the orbiter's nose. This kind of shedding had been detected only a few times, but the chunks were larger.
Until October 2002, shuttle engineers thought they had fixed the bipod foam shedding. They had not seen any occurrences in about a decade. Then, during a launching of the shuttle Atlantis, a flying fragment of bipod foam had put a serious dent in a rocket booster.
As the shuttle went though extreme temperature swings during the rush to orbit, the factory-applied foam was intended to keep the ultra-cold propellants at the right temperatures, to minimize formation of ice on the tank that could come loose and strike the orbiter, and to shield the tank from the heat of aerodynamic friction and the multi-thousand-degree burning of propellants. But the foam itself had become part of the problem.
After the October incident, shuttle managers had directed engineers to explore options to minimize or eliminate the foam shedding, but they had not halted flights. The team shared a fixed belief that the foam shedding was not a safety threat. They had always categorized it as a "maintenance" concern. That is, repairing the damage would cost time and money -- but nothing more alarming.
On the next shuttle launch, in November, no foam came off the bipod area. This made it even easier for NASA to rationalize going forward with the much-delayed launch of Columbia.
Now, with all this in their minds, on the fifth day of Columbia's mission, Don L. McCormack Jr., a senior structural engineer, gave the management team its first formal report on the foam strike: "As everyone knows, we took the hit . . . somewhere on the left wing leading edge." The review was still going on, he told Ham, and "we're talking about looking at what you can do, uh, in event we really have some damage there but . . ."
Ham interjected. "Hey, just a comment. I was just thinking that our flight rationale [for going ahead with launching after the foam strike in October was] that the material properties and density of the foam wouldn't do any damage . . ." She suggested looking at that data and also the data from a 1997 flight where there had been debris damage.
McCormack agreed and noted that on the earlier mission, "we saw some fairly significant damage area" on the wing -- but on the glassy ceramic tiles that cover the underside of the orbiter, not to the carbon fiber panels on the leading edge.
Returning to Columbia's situation, Ham continued, "And really, I don't think there is much we can do, so you know it's not really a factor during the flight . . ."
That same Tuesday in Houston, engineer Paul Shack sent an e-mail to a senior engineer named Alan R. (Rodney) Rocha, in the structural engineering division at Johnson Space Center, saying that engineers at Marshall Space Flight Center in Huntsville, Ala., concluded that the tank foam loss "over the life of the Shuttle program has never been a 'Safety of Flight' issue.' " Shack added that this did not sit well with Bryan O'Connor, NASA's top safety official and a former astronaut, who asked for a hazard assessment.
Aboard the shuttle that day, Ramon and crewmates had a space-to-ground chat with Israeli Prime Minister Ariel Sharon in Jerusalem.
Engineers Wanted Images
Also on that Tuesday, Jan. 21, a large group of Houston engineers responsible for troubleshooting the foam impact made a formal decision to seek spy satellite images of the orbiting shuttle. The area where the foam hit the wing was not visible from the shuttle's crew cabin.
Summarizing the decision in an e-mail the same day, Rocha said the participants had agreed that "big uncertainties" would remain "until we get definitive, better, clearer photos of the wing and body underside . . . Can we petition (beg) for outside agency assistance."
The next day, he heard that upper managers had denied the request. He then drafted an e-mail to 14 colleagues, had some conversations about it -- but never sent it.
"In my humble technical opinion, this is the wrong (and bordering on irresponsible) answer . . .," he said, citing potentially "grave hazards." He acknowledged that there was no certainty the images would help, but added, "Remember the NASA safety posters everywhere around site stating 'if it's not safe, say so?' Yes, it's that serious."
This was one of at least two requests for such images during the mission. One was an informal query from a shuttle official at Kennedy Space Center about enlisting the Air Force's help. Senior managers squelched the request.
As Columbia orbited, manager Ham heard in a phone chat that there had been a request for the imagery and spent most of a day trying to track down its source. She asked contractors and the key engineer groups. No one spoke up.
"I was too low down here in the organization, and she's way up here," Rocha would later tell ABC News, explaining why he never took a stand on the potential danger. "I just couldn't do it."
The story of NASA's failure to obtain these images would prove to be convoluted and complicated. For example, investigators would later find that the denial of the informal request was mistakenly taken as a denial of the formal plea from Rocha's group.
Investigators would blame the entire system as well as a large number of individuals for missed communications going up and down the chain of command as well as for allowing their knowledge of classified imaging capabilities to wither.
No one would ever know whether the images might have revealed the wing damage. But investigators and shuttle "old-timers" would say later on that was beside the point: If managers had handled the shuttle properly, as a high-risk experimental vehicle, there would have been no question about the need to try to get thorough photography and other documentation.
"Of course you want pictures, just so you know," retired Adm. Harold W. Gehman Jr., chairman of the investigating board, would tell Sen. John McCain (R-Ariz.) at a hearing.
Relying on the Boeing Team
The shuttle team was relying heavily on the work of a small team of Boeing company engineers assigned to analyze the potential damage. In a preliminary report dated Thursday, Jan. 23, the Boeing group concluded the debris had not caused enough damage to imperil vehicle or crew.
Late that day, a week after liftoff, mission control sent a private e-mail to Husband and McCool informing them of the foam impact but assuring them that it was "not even worth mentioning," except for the chance a reporter might ask them about it.
It was not until the next day, Friday, Jan. 24, that the mission managers held their most detailed discussion of the foam impact.
Engineer McCormack told the group that the ongoing Boeing analysis indicated no serious threat. He noted that the analysis had used a computer software program known as Crater. However, no one in the meeting mentioned that the program was not intended for use in in-flight decision-making.
McCormack told the managers that, at this stage, engineers "do show obviously there's potential for significant tile damage here" and for localized heat damage, but not for a complete "burn-through."
He went on, "Obviously, there is a lot of uncertainty in all this in terms of the size of the debris and where it hit and angle of incidence and it's difficult . . ."
Ham cut him off, "No burn-through means no catastrophic damage, and localized heating damage would mean a tile replacement?"
McCormack assented.
Ham recapped: no safety threat, nothing "we're going to do different," only possible between-flight repairs?
"Right, right," McCormack answered. "It could potentially hit the RCC and we don't indicate, other than possible coating damage or something, we don't see any issue . . ." A second engineer agreed.
Although Ham asked about the size, no one mentioned that the foam fragment was the biggest yet detected -- in fact almost twice the size of any seen previously.
A Flurry of E-Mails
Elsewhere in the organization, there were muffled signals of anxiety in the requests for satellite imagery and an informal consultation that had sprung up. It was initiated by mid-level mechanical systems engineers at Johnson Space Center in Houston with specialized colleagues at NASA's Langley Research Center in Hampton, Va. But these discussions would never fully surface at higher levels.
As the weekend commenced, Boeing was still assessing one last damage scenario: What if the foam had hit in the vicinity of the main landing gear door?
It was in connection with this issue that mid-level engineers in Houston had consulted the Langley engineers, specialists in landing gears and related systems.
The Boeing engineers ultimately concluded that there could be elevated temperatures reaching an outer layer of the shuttle's aluminum structure, with damage to the heat shielding and door structure, but "no predicted burn-through of the door," and no threat to safety.
That Sunday, Jan. 26, at 7:45 p.m., Rocha distributed an e-mail to more than a dozen colleagues describing the Boeing conclusions and noting that he had accepted them. "This very serious case could not be ruled out and it was a very good thing we carried it through to a finish."
Even as the Boeing analysis was ending, the exchange continued between the engineers in Texas and Virginia. In their e-mails, they outlined dire worst-case scenarios, "carnage in the wheel well," and options for dealing with them. The participants would later describe these exchanges, which continued through the mission, as standard "what-iffing" -- the sort of thought exercise engineers were encouraged to engage in as prudent preparation for possible crises. But it was also the first time Langley veterans could remember the Houston engineers consulting them this way about a shuttle mission in progress.
On Monday, Jan. 27, Langley's Robert H. Daugherty sent this message to his friend, Johnson Space Center engineer Carlisle Campbell:
"WOW!!! I bet there are a few pucker strings pulled tight around there! Thinking about a belly landing versus bailout . . ."
That same Monday, by contrast, the main landing gear door scenario was barely mentioned in Linda Ham's management meeting. McCormack summed up the Boeing assessment of no serious danger.
"A turn-around issue?" Ham asked, referring again to between-flight repairs.
"Yeah, possibly," the engineer replied.
On Tuesday, Jan. 28, Daugherty at Langley wrote his friend at the Houston space center again to ask, "Any more activity today on the tile damage or are people just relegated to crossing their fingers and hoping for the best?"
Shuttle managers formally approved the Boeing analysis that day -- the 17th anniversary of the Challenger disaster. A copy of the findings went to NASA headquarters in Washington.
At no point would any of those involved in the "what-iffing" express special concern to higher-ups who could take action. Upper management's closest known hint that the discussion was even going on was contained in an e-mail that Campbell, in Houston, sent to top safety official O'Connor in Washington. The two had worked together years earlier.
"Bryan, this is confidential, but I just wanted to be sure that you were aware of the potential landing gear door damage," Campbell wrote on Thursday, Jan. 30. He included Rocha's e-mailed description of the analysis and conclusions as approved the preceding Sunday: a possible gouge but no threat to safety.
"Thanks for the info," O'Connor responded the next day, Friday. "I had heard a little bit of this but did not realize so much analysis was required . . ."
The Crater program's predictions would turn out to be correct about the nonlethal effects of the foam striking glassy tiles. But later, in painful hindsight, a glaring misstep in the engineering calculus would become clear: The team had assumed that the tile analysis told them all they needed to know about the potential damage to the very different RCC material as well.
Conventional wisdom among the engineers was that the RCC, designed to withstand higher temperatures than the tiles, was also more resistant to impact damage. But they really did not know. Nobody had tested the question. This fact had been clearly noted in Boeing's written Jan. 23 assessment of the potential damage to Columbia: "No SOFI [spray on foam insulation] on RCC test data available."
The engineers had, in effect, been guessing. And neither Ham nor any other manager challenged the conclusion.
On Wednesday, Jan. 29, the lingering issue of a spy satellite inspection of the shuttle surfaced at NASA headquarters, when safety official Michael Card approached William Readdy, head of space flight and a former astronaut, to pass on an imaging agency's offer to take such pictures. Readdy, convinced by mission managers that the shuttle was in no danger, sent word that the agency need not consider the images a high priority.
On Thursday, Jan. 30, with almost two days to go in the 16-day mission, Ham's managers met for the fifth and final time. The only discussion of the foam impact addressed engineers' eagerness to get hold of the astronauts' launch-day photography of the tank, which might show the scar where the foam had come off.
On Friday, Jan. 31, flight director Leroy Cain briefed reporters about the possible debris damage. Engineers, he said, "took a very thorough look at the situation . . . and we have no concerns whatsoever."
Reentry Begins
Just past 8:15 a.m. the next day, Husband and McCool, flying over the Indian Ocean, fired Columbia's braking rockets for two minutes and 38 seconds to begin the hour-long descent. Columbia would fall for about half an hour before it hit the first traces of atmosphere 400,000 feet above the Pacific Ocean, northwest of Hawaii.
After flying 6.6-million miles in 16 days, they were a mere 5,113 miles from touchdown. It was just past 8:43 a.m. EST. They talked in fascination about the fiery plasma -- glowing, superheated air -- the hypersonic shuttle was beginning to generate around them.
McCool said, "That might be some plasma now."
Clark: "Think so, already?"
McCool: "Yeah, . . . We see it out the front, also."
At 8:44 a.m., still going almost 25 times the speed of sound, the shuttle officially started "feeling" the atmosphere.
McCool: "It's going pretty good, now. Ilan, it's really neat, just a bright orange yellow out over the nose, all around the nose."
Husband: "Wait until you start seeing the swirl patterns out your left and right windows."
McCool: "Wow."
Husband: "Looks like a blast furnace."
In mission control, the flight team was discussing the weather at Cape Canaveral for the landing.
McCool: "This is amazing, it's really getting, uh, fairly bright out there."
Husband: "Yep. Yeah, you definitely don't want to be outside now."
Chawla: "What, like we did before?" The astronauts laughed.
'I Am Not Believing This'
Within five minutes after the orbiter's first taste of extremely thin atmosphere, the superheated air had begun to penetrate the wing.
The ship was flying with wings level and nose angled up about 40 degrees from horizontal, an orientation designed to keep the black heat-shield tiles that covered the underside facing the brunt of the heat. The RCC-covered nose and wing leading edges typically experienced the highest temperatures -- up to 3,000 degrees Fahrenheit.
The hole from the foam impact was on or near the underside of Panel 8 of 22 RCC panels, at the point where the sweep of the wing angled outward. About 8:48 a.m., a sensor near RCC Panel 9 showed the first unusual heating. Searing air was eating into the aluminum framework of the wing.
Columbia was approaching the continental United States, about 48 miles high, going more than 16,000 mph. As it metamorphosed from spacecraft to aircraft in the building air pressure, it became a glider that used the drag of the atmosphere to slow it for landing. The ship began the first in a series of four planned steep banking maneuvers to help it slow down, this one to the right.
At this moment came the first of more than a dozen unusual interruptions in radio communications. Investigators would suggest later that the signals were blocked by the molten aluminum and other metals from the failing wing that were spraying into the surrounding air, with an effect similar to that of "chaff" used by military planes to confuse enemy radar.
Normally, as the shuttle dropped into thicker atmosphere, the smooth flow of air over the surface would act as a kind of insulator, keeping the hottest part of the air at a distance. With the breach in Columbia's wing, however, that smooth flow was disrupted, exposing some surfaces to much more intense heat than they were designed for -- possibly as much as 8,000 degrees. near the hole. (The aluminum support structure would melt at 1,200 degrees.) The prolonged, extreme heating dramatically eroded the RCC material, normally half an inch thick, to a knife edge.
About 8:52 a.m., less than eight minutes after it had entered the atmosphere, Columbia's computerized flight control system started compensating for increasing drag on the ruined left wing. At this point, the searing gases burned all the way through the wing's aluminum support and flowed into the wing's hollow interior, spreading with remarkable speed. The intruding heat plume reached a part of the main landing gear wheel well enclosure and started burning through bundles of electrical wiring, one after another, from top to bottom. Over about another minute, 165 sensors would fail.
About this time came the first abnormal indicator -- a small rise in a wheel-well temperature sensor -- that might have been seen by the crew or ground team in "real time."
Just after 8:53 a.m. EST, Columbia crossed the coast of California north of San Francisco. Although it was still an hour before sunrise there, a few sky-gazers turned out to glimpse the fireball streaking overhead. Within 20 seconds, people on the ground were stunned to see multiple flashes of debris coming off the orbiter -- possibly shards of upper wing skin and bits of heat-shielding.
At just past 8:54 a.m., as the orbiter crossed into Nevada, mechanical systems officer Jeff Kling in Houston, monitoring shuttle telemetry, told flight director Leroy Cain, "I've just lost four separate temperature transducers on the left side of the vehicle, hydraulic return temperatures."
Cain, who had worked a dozen or so shuttle missions, instantly thought of the foam impact.
"Okay, is there anything common to them?" Cain responded. ". . . I mean, you're telling me you lost them all at exactly the same time?"
Shortly after 8:54 a.m., the shuttle's guidance system started fighting to keep the orbiter stable, possibly because of a sudden deformity in the wing as its frame melted.
About this time, video-equipped witnesses on the ground saw debris falling away and odd flashes -- including a notably bright one that lighted up the already glowing sheath of air around the shuttle for several seconds. These could have been pieces of upper wing surface, heat-shielding or a fountain of molten metal spurting into the outside air. Ten more pieces would come off over the next minute.
Kling responded to Cain, "No, not exactly. They were within probably four or five seconds of each other . . ."
About 8:56, just as Columbia moved from darkness into sunlight, the superheated gases finally burned their way into the wheel well itself, sending temperature readings there soaring.
Even though it was starting to come apart, the orbiter now began a normal, planned roll to the left. An Air Force telephoto view of the Columbia as it streaked over New Mexico, later widely publicized, revealed abnormal patterns in the airflow ahead of and behind the left wing..
At 8:58 a.m., Cain asked Kling about the sensor measurements, "And when you say you lost these, are you saying that they went to zero . . .?"
"All four of them are off-scale low" -- no longer functioning.
The shuttle shed more pieces -- possibly large segments of the left wing.
Cain asked a guidance and navigation officer, "Everything look good to you, control and rates and everything is nominal, right?"
The officer responded, "Control's been stable through the rolls that we've done so far, Flight. . . . I don't see anything out of the ordinary."
Just after 8:58 a.m., as Columbia crossed into Texas, Husband radioed from the orbiter, "And, uh, Hou[ston] . . ." The transmission was garbled.
Thirty seconds later in mission control, Kling had worse news for Cain, "We just lost tire pressure on the left outboard and left inboard, both tires."
Astronaut Charles Hobaugh, talking to the crew from mission control, said, "And Columbia, Houston, we see your tire pressure messages and we did not copy your last."
Cain asked Kling, "Is it instrumentation . . .? Gotta be . . ."
"Those are also off, off-scale low," Kling said. In an aside to a colleague, he added, "I am not believing this."
Just before 9 a.m., all incoming telemetry suddenly disappeared from the screens in mission control. A final brief burst of data would reach ground computers a bit later. The onboard data recorder would keep functioning until the vehicle came apart and sent it hurtling into an East Texas slope.
As Columbia streaked through the thin atmosphere more than 200,000 feet high, approaching Dallas at 18 times the speed of sound, the shuttle fired four right jets and adjusted trim dramatically to compensate for a tremendous force pulling it to the left.
In the final transmission from the doomed ship, at 8:59:32 a.m., Husband replied to Hobaugh, "Roger, uh, buh . . ."
Thirteen seconds after that, a sizable portion -- possibly all -- of the left wing was gone. Over the next 30 seconds, five more pieces flew off accompanied by two flashes of light.
The ship finally went out of control. The ship had lost its cooling system, but the electrical power and life support system were operating. Multiple alarms sounded. Husband or McCool apparently bumped the joystick, but there was no sign they intended to take control away from the auto pilot.
Columbia's main fuselage started disintegrating at 18 seconds past 9 a.m. EST, 46 seconds after Husband's final broken message.
No one knows how long the astronauts were conscious to consider their fate. It is possible that, as in the case of Challenger, the reinforced crew cabin remained intact some seconds longer than the rest of the vehicle. Then, still hypersonic, it too was pulled apart as it hurtled through the thickening air.
In mission control, the flight team tried repeatedly to reach the crew or pick up some sign of the vehicle.
Twelve agonized minutes after Columbia's breakup, missions operations representative Philip L. Englauf took a phone call informing him that a Dallas TV station had video of multiple objects trailing vapor along Columbia's flight path. Englauf leaned over his console and told Cain, who sat in front of him.
Cain's shoulders sagged, and he paused to pray for the crew and their families. He fought back tears as he turned to face about 20 people and instruct them to get to work on the disaster plan.
He said, "Lock the doors."
© 2003 The Washington Post Company
|
|
