The future of NASA's three remaining space shuttles may now hang in part on deciphering an artifact from the past, a 1970s vintage reel-to-reel magnetic tape.
This technological dinosaur is for the moment the most celebrated piece of forensic evidence among the almost 50,000 fragments collected so far from the wreckage of the shuttle Columbia, which are accumulating in hangars at Kennedy Space Center.
Somewhere on the recorder's 9,400 feet of 1-inch-wide, 28-track tape, on reels the size of large pizzas, investigators hope to find measurements of the strains, stresses, vibrations, temperatures and pressures that buffeted Columbia's left wing, where the doomed shuttle's troubles began. If the recorded data survived the heat and trauma of its plunge to a muddy hillside near Hemphill, Tex., they should span the entire reentry until seconds before the orbiter's final catastrophic breakup over Texas on the morning of Feb. 1.
"In a perfect world, we will have 721 measurements" from sensors scattered around the wings, fuselage and vertical tail surfaces of the space plane, said G. Scott Hubbard, a member of the Columbia Accident Investigation Board. These would include 182 pressure measurements, 53 temperature measurements and 447 measurements of dynamic stresses, he said. The recorder was installed to monitor the performance of Columbia, the first shuttle, in its initial test flights. It supplements the telemetry that streams continuously to mission control.
Technicians were working to clean, copy and analyze the surviving data this weekend after the flight data recorder was sent on an elaborate round trip from Texas to Minnesota to the NASA launch complex here in Florida and then back to Texas, carried on a T-38 astronaut training jet as one special team handed it off to the next.
It almost did not happen at all.
On their first sweep of the area where the recorder was found, searchers walking a grid across the Texas piney woods and fields did not notice the partially buried prize. They were sent back for a second try after investigators in charge of reconstructing the wreckage were able to figure out where the box should have fallen -- a fact that reflects the growing sophistication as well as the daunting complexities of the unprecedented operation.
The team hopes this technique will help locate other key missing pieces, notably two cameras carried in the belly of the orbiter that recorded a key sequence during Columbia's Jan. 16 ascent.
Although the tape may be deciphered within days, officials said, engineers and an array of specialists are just beginning what promises to be a methodical, months-long effort to analyze other telltale elements of the charred and twisted evidence. If the story the physical evidence tells conforms with the telemetry data, imagery and other lines of inquiry, investigators will be confident that they have pinned down the probable sequence of events that led to the breakup, which killed seven astronauts. That, in turn, will tell them how to fix the problems and clear the way for the shuttles to fly again.
"We're doing our very best, but this is the first time in human history anyone's had to analyze a hypersonic space reentry disintegration of a craft," said Gregory T. R. Kovacs, of Stanford University, a specialist in electrical and biotech engineering who is assisting the investigation. "So there's no one to ask how long will it take."
Preliminary evidence indicates that a breach on or near the leading edge of the left wing allowed a plume of superheated air to burn through the shuttle's skin as it plunged back into Earth's atmosphere. The breach may have opened at a point weakened by the impact of one or more chunks of foam insulation that detached from the shuttle's external propellant tank during launch.
The shattered remnants of Columbia -- about 26 percent of it by dry weight, so far -- continue to arrive at Cape Canaveral at the rate of two truckloads a week, each containing as much as 4,000 pieces of wreckage. The main hangar where the reconstruction effort is underway sits next to the shuttle landing strip, within about 2,500 feet of the spot where the orbiter was aiming for a touchdown that Saturday morning almost two months ago. Using blue and yellow tape, technicians have traced a loose outline of the orbiter on the 60,000-square-foot concrete floor, where they lay out significant recovered parts in a grim approximation of the original.
The orbiter's demise was so violent that the largest piece recovered so far is a piece of its belly about 14 by 5 feet, and most are much smaller, said Michael Leinbach, the Space Center's launch director and currently chairman of the reconstruction effort.
Most of the aluminum frame and skin of the 90-ton orbiter have disappeared, melted and vaporized by the heat of the calamitous reentry.
"A lot of the parts have come through an aluminum molten rain cloud, if you will," said Steve Altemus, a shuttle test director working on the reconstruction.
Two shifts of 75 workers each are working six days a week to identify as many parts as possible. They create a "data pack" for each piece, which includes photos and a bar code that allows them to track the item in a computer data base for the life of the project.
Many pieces -- tiles, tubing, wire -- are unrecognizable, Leinbach said. Others have been deemed irrelevant to the investigation. Out of sensitivity for the feelings of the crew's families, parts from the crew cabin are walled off out of sight in a corner of the facility. All in all, only about 1,400 of almost 50,000 pieces have been laid out on the 40,000-square-foot grid, on or near their proper place on the outline of the orbiter.
Specialists have launched structural, chemical and metallurgical studies of the remnants in hopes of translating complicated damage patterns into a coherent narrative of what happened. Their tools include examination with stereoscopic microscopes, X-rays and ultrasound, fracture analysis and electronic mapping with color coding to show where each part belongs on the orbiter. Laser scanners create a virtual computer model of the wreckage, which they hope will help them mate fractured surfaces and recognize patterns that cross multiple pieces of wreckage, Altemus said. About 140 parts have been scanned in to date.
"What we're trying to do is get an idea from the flow patterns, the splatter patterns and the deposits, where was the breach, what was the origin this whole thing started from," G. Mark Tanner, a forensic analysis expert assisting the investigation, told the board.
One of the difficulties is sorting out the sequence of events. "A key question when we're looking at these pieces is . . . was [the damage] caused by something that happened on ascent, on descent, at breakup, or from ground impact," Kovacs said. Some pieces, he noted, had "pine needles embedded in them. So clearly, that did not happen in space. We're taking great care to understand the relationships."
He compared it to "putting together a multi-thousand-piece 3-D jigsaw puzzle on a 2-D surface."
In the search for the flight data recorder, the reconstruction team used the data base to establish where pieces from the same section of the crew cabin had landed, Leinbach said. The process was complicated by the fact that fragments of different shape and density had different aerodynamic properties causing some to "fly" farther before they fell to Earth, and some of the wreckage probably broke into smaller and smaller pieces on the way down.
But it worked. "The information was passed on to the searchers. We got them down to an area maybe about the size of this hangar, to go search again," Leinbach said. "It's kind of easy to understand how they missed it the first time," because the recorder was half buried in soft ground.
"Now we know it's a proven technique," he said, they intend to focus searchers next on two cameras, a 35mm still camera and a 16mm video camera, that were mounted on Columbia's underbody where it attaches to the external tank -- an area from which the foam debris appeared to come loose during Columbia's launch.
The cameras photographed the tank as it separated from the shuttle shortly after reaching orbit, he said, and may reveal crucial information about the foam material that broke free during launch.
The search effort along a 240-mile swath of Texas and Louisiana, now about two-thirds finished, is expected to go on until early to mid-May, when spring foliage will make the effort much more difficult, officials have told the board. "There are 80 folks, 10 feet apart, and so an 800-foot line of Forest Services rangers walking the grids for us," Leinbach said.
The search, augmented by low-flying helicopters, is not without its own hazards. On Thursday, a U.S. Forest Service helicopter crashed in an East Texas forest, killing two people.