13 October 2004 – Col Dave Nakayama presents Columbia investigation findings 13 Members of Columbia Accident Investigation Board (CAIB) 192 Staff, 40 outside consultants Group 1- Materials Group 2- Operations Group 3 -Engineering/Technical analysis Group 4 - Organization and Policy Independent Analysis Team By weight, 38% of the Orbiter was recovered. 30,000 documents were reviewed. 200 interviews were conducted. 84,000 pieces of wreckage were recovered. 3000 “fault line” elements had to be investigated until they could narrow it down to a single cause. 3000 public inputs were added. Site visits took place at Johnson Space Center (Pasadena, TX), Kennedy Space Center, Marshall Space Center (AL), Michoud Assembly Facility, Stennis Space Center (MS), Palmdale, CA, and ATK Thiokol in UT (the manufacturers of the Booster rockets which propel the Shuttle into orbit; they were also implicated, as Morton Thiokol, in the 1986 Challenger disaster). The mission of the board was simple: identify/prove the physical cause of the accident. The prime suspect: breach in the Thermal Protection (the tiles and other re-entry safe portions of the Orbiter which protect it during re-entry). The cause was an impact of foam from the external fuel tank which occurred 81.7 seconds into the launch. Once the mission was accomplished of diagnosing the problem, the layers needed to be peeled back in order to avert the next disaster. So why do you want maintenance officers on this board? Because: the Shuttle was old, they could look more closely at issues such as Initial Certification/Service Life Extension, management issues, and above all else, provide leadership! The Shuttle is ancient. 1 The Columbia, first launched on 12 April 1981, was on mission #28 when it broke apart. A Shuttle has 2.5 million parts and weighs 4.5 million pounds. 7 million lbs of thrust. 2 It accelerates to 17, 400 miles per hour in 8.5 minutes. Engine temps can range from –420 to 6000 degrees Fahrenheit. Re-entry requires deceleration from Mach 25 to 200 miles per hour. Reentry temperatures can hit 10,000 degrees F. 24,000 tiles and 2300 flexible blankets protect the Orbiter from the heat of reentry. To service the Shuttle after
Even pilots and mission commanders would agree. The Shuttle was designed in the early 1970s and to this day is equipped with avionics dating from this period. A pilot, for instance, who is used to planes like the 777 or the F-16 with its modern avionics and conveniences such as heads up displays (HUDs), color monitors and digital readouts would be shocked to step into a Shuttle cockpit which still uses analog displays and monochrome monitors. 2 This is roughly the same amount of thrust that a single Saturn V rocket, which took the Apollo astronauts to the moon, can deliver.
1
13 October 2004 – Col Dave Nakayama presents Columbia investigation findings a mission, 1.5 million work hours are required and 73, 500 inspection points must be looked at. By way of comparison, a Boeing 747 requires 22,000 work hours and 1800 inspection points. On liftoff on 16 January 2003, the debris hit reinforced carbon-carbon panel #8, creating the 6 to 10 inch diameter breach of the leading edge. The wing cooked at 5000 degrees and from there it took 15 minutes for the shuttle to disintegrate during the morning of 1 February 2003. None of the astronauts aboard knew what hit them. So where specifically did the foam come from? It was used to insulate mounting hardware, the hard points where the Orbiter is merged with the External Fuel Tank. This piece weighed a mere 1.67 pounds, but since it was traveling at 528 miles per hour at impact, this was a devastating impact. Imagine a basketball thrown at 500 miles per hour, or a bag of cement thrown at 70 miles per hour. Original design requirement: NO FOAM LOSS from external fuel tank!!! BUT, from the word go on 12 April 1981, foam loss was repeatedly experienced. 300 tiles had to be replaced. 3 Photographic evidence of foam loss that could or did cause the loss of tiles exists on 69 of the 75 flights. This has never been solved and accepted as normal (or what they liked to call an “in-family” anomaly). MX Issues -No standard safety and mission assurance in NASA centers -KSC and contractors S&MA workforce methodology changed significantly in the 1990s. Inspections were cut in half from 1989 to 1996. -KSC management perceived as unsupportive of QA -FOD program took a wrong turn: “looking good” became the goal instead of results. Many decisions were budget driven. Palmdale, CA Orbiter major mod killed/consolidated with KSC in 2001. That decision drove a higher ops tempo: 4 orbiters, 3 OPFs. This demanded better integrated planning, scheduling but this did not happen. Hence more things began to fall through the cracks. Work Authorization Documents (WADs) contain tech data derivatives for individual tasks. WADs require documentation of the completion. CAIB got all of this for STS-107 (the Columbia mission) and STS-109 (another mission to occur after this but was canceled) 99% correct…out of 600,000 documents. This means that although this is spoken of in terms of just 1%, that 1% accounts for a lot of
I remember reading books on the Shuttle as an elementary school student in the 1980s which mentioned that “the Flying Brickyard”, as it was sometimes called, routinely lost tiles, but that the loss of tiles was just that – routine.
3
13 October 2004 – Col Dave Nakayama presents Columbia investigation findings pages! That 1% also included critical misses. These WAD errors impacted CoFRs (Shuttle-lingo for an “exceptional release”). There also existed a backlog of Unincorporated Engineering Orders. Over 1400 of these UEOs had 710 changes each! Past efforts to rectify this problem always fell short in terms of funding. Priorities were misaligned, The basic subsystem life for one Orbiter is 100 times in 10 years 4. In 1991 they were recertified for another 10 years. Originally programmed for replacement in early 2000s. But NASA changed plans to fly the Orbiter until 2020. NASA forced to look at structural fatigue, corrosion analysis, Kapton wiring. Mid-life recertification was being explored. All of the infrastructure used for the Shuttle, to include the Vehicle Assembly Building, the “crawler” which moves the Shuttle assembly from the VAB to the launch pad, to Launch Pad 39B itself, all date from the Apollo era. This is not to say that NASA wasn’t concerned about safety. There was a device for the Shuttle called the Separation Bolt Catcher, which was supposed to catch debris falling from SRB separation. Routinely, fragments leave SRB and EFT upon severing from the Orbiter. But the Bolt Catcher in its present configuration never qualified. What if it had actually worked…or what if it had failed? Each orbiter has 852,000 feet (161.36 miles) of Kapton wiring. 2000 feet of this is inaccessible! The major concern here is arc tracking. Is everything an engineering problem? Nah. 4 of the 10 volumes produced by the CAIB are aimed at history/organization/culture as factors in the loss of Columbia.5 History -NASA needed a new system -Impacted by the end of the Cold War
If the Challenger accident had not happened, there would have been 24 Shuttle launches in 1986. Such were the lofty plans for that year. This would have amounted to a Shuttle round-the-clock in orbit, and indeed, as if to set the stage for that year, the only time two Shuttles appeared together on the launch pads ready to go was in January 1986, when the Columbia was launched on 12 January to inaugurate the new year, and the ill-fated Challenger was pushed back from the original launch date in late 1985. 5 According to CBC’s news program The National, one particular one which aired in late August 2003, NASA suffered from “a culture of risk-taking.” This line was perhaps quoted directly from the CAIB’s findings.
4
13 October 2004 – Col Dave Nakayama presents Columbia investigation findings -Hamstrung by politics and dwindling budgets, driven to “faster, better, cheaper” concept -The Shuttle was mischaracterized as a mature, reliable operational system. What the Shuttle was, and is, is an experimental craft and nothing more. Organization No independent safety organizations NASA safety offices have too many blind spots These offices were also rife with barriers to communication Culture A closed society/culture Many post-Challenger reforms Cultural resistance to externally imposed change As a result, there was reversion. 2003 was essentially 1986 for NASA. Santayana’s Law “He who ignores history is condemned to repeat it” was not followed at NASA. -Schedule pressures -Silent Safety program -Lack of critical deviations Suggested reading: “Beyond The Widget” by BGen Deal, Air & Space Journal, 2004. Bottom line: technical causes not everything!