Follow us on

|

WHAT WENT WRONG?

Print edition : Feb 28, 2003 T+T-
Columbia breaks up in the sky over Tyler, Texas, on February 1, as seen in one of a series of pictures shot by an amateur photographer.-DR. SCOTT LIEBERMAN/AP

Columbia breaks up in the sky over Tyler, Texas, on February 1, as seen in one of a series of pictures shot by an amateur photographer.-DR. SCOTT LIEBERMAN/AP

Could the seven astronauts on-board Columbia have been saved by assessing the debris-hit correctly and using the International Space Station for a rescue mission? A shocked world waits for answers as investigators try to unravel the mystery behind the catastrophic break-up of the space shuttle.

AT a press briefing the day before the scheduled landing of space shuttle Columbia, the entry flight director, Leroy Cain, said: "We did our normal flight control systems check-out... all the systems we use for entry. The vehicle performed flawlessly today as it has during the entire mission." He, however, added: "The only issue is a bit of possible damage on Columbia's left wing. Video clips of the launch shows what appears to be a piece of foam insulation from the shuttle's external tank falling away during ascent and hitting the left wing near its leading edge. But it is not significant." But after disaster struck Columbia when it was returning from its science-driven mission STS-107 on February 1, he must have wished that he had not uttered the last sentence. For, it is the anomalous temperature rise in the left wing, resulting from a failure of the protective heat shield, that seems to have led to the catastrophe.

A January 17 report of the post-launch high-speed film review had said the following: "At approximately 80-84 seconds after T-0, a large piece of debris is observed striking the underside of the LH [left hand] wing of the orbiter. The debris appears to originate from the area of the bipod attach point on the external tank. No damage to orbiter Thermal Protection System (TPS) is apparent." At 33 seconds after lift-off (T-0), the film showed a shower of particles from the portion below the left solid rocket booster, which were concluded to be pieces of foam insulation from the attach point of the external tank that carries liquid hydrogen and liquid oxygen fuel. The rigid foam insulation can peel off because the tank contracts when filled with liquids at ultracold temperatures.

All available evidence at the moment seems to indicate that the above event may be the root cause of the mishap, with the orbiter's thermal protection tiles on the left wing having been damaged by the impact of the debris.

A report two days later said analysis of the films could not identify individual tiles, but concluded that no large-scale damage had occurred. On January 28, a report of further analysis said that the impact could have damaged a swath of tiles as large as 18 cm wide and 80 cm long. But it concluded that there was "no safety of flight issue". Thus, despite being aware of the falling debris at lift-off, NASA engineers had ruled out any possibility of damage that could be fatal to the orbiter. "Engineers took a thorough look at the situation with the tile on the left wing and we have no concerns whatsoever. We have not changed anything with respect to our trajectory. It will be nominal, standard trajectory," Cain had said the day before the tragedy. But it was not to be. Columbia's trajectory veered out of control towards doom.

In a nominal shuttle re-entry scenario, the re-entry phase starts about one hour before touchdown and at some 8,000 km from the landing site at the Kennedy Space Centre (KSC) in Florida. At that point, the shuttle is at an altitude of about 170 km and is travelling at about 28,000 km/h. It is put into its correct orientation and reaction control system jets are fired to start the descent. About five minutes later, at an altitude of about 120 km, entry into the upper atmosphere begins with an automatic sequence monitored by an on-board control system.

The whole re-entry phase can either be flown automatically or under crew control, but not much can be done manually because once the descent starts the orbiter is in glide mode. The re-entry on February 1 was not nominal. At about 14-00 GMT, communication with the space shuttle was lost. At this time, the shuttle was flying over Texas, about 15 minutes from landing and at about 18 times the speed of sound (March 18, about 20,000 km/hr). It was at an altitude of 63 km, about 1,400 km from the landing site at KSC. Amateur video films show the spacecraft following a constant track and apparently disintegrating slowly.

During re-entry, atmospheric friction causes the orbiter to be subjected to enormous heat loads with temperatures rising up to 2,000{+o}C as it plunges into the atmosphere. The shuttle is, therefore, protected by a heat shield or TPS that is composed of reinforced carbon-carbon composites in the hottest areas (the nose cone and the edges of the wings), protective blankets in the coolest areas (mostly the top side of the orbiter) and silicate tiles (about 25,000 of them, each measuring 20 cm x 20 cm x 5 cm) covered with black glazing and bonded to the underbelly of the vehicle. Like a normal aircraft, the shuttle too comes down with a large angle of attack, with its nose up and the face of the underbelly meeting the atmosphere in its approach to landing. The angle of attack - which is usually 40-45 - is critical because a larger angle would expose the more vulnerable parts to the heat build-up and, if it is less, the entry would be too fast again leading to greater heat loads. The abnormal events indicated by telemetry data point to a slow penetration of re-entry heat into the orbiter owing to failure of the TPS on the left wing side, causing drag on the vehicle towards left. The system failure and anomalies that followed overpowered the on-board flight control system (FCS).

THE first sign of a significant anomalous behaviour in an otherwise flawless mission was telemetred about eight minutes before the craft broke up (graphic, `Columbia's last minutes', on page 6). It was an unusual rise in temperature in the left wheel well area. "This was the first occurrence of a significant thermal event," according to Ron Dittemore, NASA's Shuttle Programme Manager. "Engineers," he said at the post-crash briefing, "do not believe the left wheel well was breached, but rather that hot gases were somehow finding a flow path within the wing to reach the wheel well." Insulation pieces falling off is not uncommon in shuttle launches but this time the chunk seems to be the biggest ever.

According to Dittemore, the tank debris measured about 50 cm x 40 cm x 15 cm and weighed about 1.2 kg. The air velocity between the shuttle and the tank was about 1,600 km/hr. The conclusion that the event was not significant was based on STS-107 images and a similar incident that occurred on the shuttle Atlantis in October 2002 (STS-112). On that flight, the crew had filmed the external tank as it separated from the orbiter after launch. The conclusion at that time was that the debris did not pose a threat to the safety of the crew or the vehicle. The analysis will be repeated once again, assuming the most conservative dimensions and weight of the debris and using modelling software that predicts the extent of damage. During re-entry, the flow of gases shifts from laminar (that is, taking place along constant stream lines) to turbulent; the later this shift happens the lesser the heat load. According to experts in aerodynamics, even the slightest roughness or damage to the heat shield tiles could cause turbulence, leading to anomalous heat build-up.

NASA administrator Sean O'Keefe has established an Inter-Agency Mishap Investigation Board, which is charged with determining the cause of the shuttle loss. Until such time the Board comes out with its verdict, with no other anomalous event having been recorded or seen, the falling debris from the fuel tank remains the prime suspect for the tragedy. The impact of this 1 kg piece of insulation may have damaged the highly fragile tiles of the TPS causing a burn-through and structural failure.

NASA officials, however, continue to believe that this is quite unlikely. They believe that the scenario is somewhat implausible. "There is some other event, there is some other missing link that we do not have yet, that is contributing to this temperature increase. It is a mystery to us," Dittemore has said. "Right now, it just does not make sense to us that a piece of foam debris could be the root cause. It's got to be another reason." According to him, with an outside temperature of 2,000, an increase of about 20-30 in the wheel well area does not point to any large-scale thermal incursion. "That's telling something else. There is some other event that is contributing to this 20-30 temperature increase on five sensors," he has argued.

Meanwhile, a report in an aviation magazine on February 7 has lent further strength to the falling debris hypothesis. According to the report, high-resolution images taken from a ground-based Air Force tracking camera in southwestern U.S. showed serious structural damage to the leading edge of the orbiter's left wing. The photo was taken 60 seconds before the vehicle broke up over Texas. The image analysis - which was apparently carried out at the Johnson Space Centre - shows a jagged edge on the left inboard wing structure close to where the wing joins the fuselage. The analysis also seems to show the orbiter's right side yaw thrusters firing, trying to correct the vehicle's drag towards the left. The fuselage and the right wing appear normal.

The report says that, according to the analysis, the jaggedness on the left leading edge, indicated that a small structural breach - such as a crack - occurred, allowing the re-entry heating to erode additional structure there. It is even possible that a small portion of the leading edge fell off at that location. The impact of foam debris may well have been responsible for the damage to the left wing leading edge.

Another key factor is that the TPS in the shuttle wing leading edge region does not have tiles but a different protective system made of carbon-carbon fibre, which is bolted on and not glued as the tiles are. To provide aerodynamic shape, each wing is fitted with 22 U-shaped reinforced carbon-carbon (RCC) leading edge structures. This implies that in addition to the possible failure of the black tile at the wing-fuselage interface area, a failure involving a weakened bolting mechanism could have exposed the leading edge to excessive heat loads. Any breach of this leading-edge material could potentially lead to a catastrophe. A large portion of the leading edge of the orbiter's wing was found near Fort Worth, but it was not clear whether it was from the right wing or the damaged left wing.

NASA officials have discounted the magazine's report saying that the photo released was very poor in resolution and one could draw many conclusions. The apparent jaggedness of the leading edge, according to them, is due to poor resolution of the image, which is a very long-range optical photo.

Meanwhile, observations of the shuttle by a military radar system has thrown up yet another piece of unusual signature. Data from the radar indicate that on January 17, roughly 24 hours after lift-off, an object separated suddenly from the shuttle, with a velocity of about 5m/sec. This has prompted NASA to pore over the telemetry data to look for any unusual event, including impact of high-velocity debris that could explain the radar signature. Apparently waste water is routinely dumped overboard by the crew, and this instantly turns into expanding clouds of ice crystals. Whether this or blowing off of large chunks of ice, which form in the water-dump nozzles, contributed to this signature is also being studied.

But most importantly, it is possible that impact by space debris could have damaged the heat-shield tiles or knocked off a piece of the the carbon-carbon panels on the wing's leading edge. Among the various possibilities that NASA is looking into, the impact of space debris is now likely to figure prominently.

The result of the various analyses and investigations may or may not ultimately identify the cause as damage to the TPS tiles from falling tank insulation debris. But the potential risk of a serious mishap due to this has been known for some years. A 1994 study titled `Risk Management for the Tiles of the Space Shuttle', by M. Elisabeth Pate-Cornell of Stanford University and Paul S. Fischbeck of Carnegie Mellon University, carried out a probabilistic risk analysis (PRA) of the risk involved based on the data of tiles damaged in 33 launches. The damage in these ranged from a low of 52 tiles to a high of 707. "The vast majority of this damage was minor but on several flights, over 200 tiles had suffered major hits (those with more than 1 sq. in of damaged area)."

Tile failures could be attributed to two causes: impact of debris of falling insulation and weak bonding with the vehicle frame. According to their numerical analysis, there was a one-in-a-thousand chance of losing an orbiter due to tile-related structural problems. Of these, 40 per cent of the probability was attributable to insulation debris and 60 per cent to debonding of the tile structure. They found that 85 per cent of the risk can be attributed to 15 per cent of the tiles. Dividing the tiled surface of the underbelly into different zones, the authors prepared a `risk-criticality' map of the shuttle.

Since this study was commissioned by NASA in 1990, the agency is well aware of the potential risk to the protective tiles. However, NASA itself does not seem to think that the risk of an accident due to damaged tiles is high. Several safety upgrades to the shuttle have been proposed and they are currently in different stages of development. While the upgrades include modifications to protect tiles and leading edges of the wings from even micrometeoroids and orbit debris, there is none addressing protection of the tiles from as simple a cause as falling debris.

IF the risk posed by the debris-hit at launch had indeed been gauged correctly while the shuttle was in orbit, could anything have been done to save the crew? Given NASA's assessment of the event, naturally no request was made of the U.S. intelligence services for the underbelly of the shuttle to be examined by reconnaissance satellites or ground-based cameras. In any case, there were no means on-board for examination or repair of any damage to the tiles of the heat shield.

However, in principle, if the damage to the tiles of the thermal shield was assessed to be serious, the `space infrastructure' of the International Space Station (ISS) could have been used for rescue operations, according to Mark Wade, a space expert who maintains an on-line space news service. Columbia could have been flown in an orbit that allowed it to rendezvous with the ISS in an emergency (albeit with reduced payload). The ISS crew could have inspected the shuttle prior to docking with the station. If the damage was serious, the shuttle could have remained docked to the station and the crew could have waited for a repair or rescue mission, Wade has argued.