Monday, January 28, 2013

T+1:02............PLT..... Thirty-five thousand going through one point five.

(NASA: Altitude and velocity report, 35,000 ft., 1.5 Mach).

T+1:05............CDR..... Reading four eighty six on mine.

(NASA: Routine airspeed indicator check.)

T+1:07............PLT..... Yep, that's what I've got, too.

T+1:10............CDR..... Roger, go at throttle up.

(NASA: SSME at 104 percent.)

T+1:13............PLT..... Uh oh.

T+1:13.......................LOSS OF ALL DATA.

On January 28, 1986, the Space Shuttle Challenger suffered a catastrophic failure 73 seconds into its flight when flames from its damaged solid rocket booster caused the main external fuel tank to explode. Contrary to popular belief, the shuttle itself was not destroyed by the explosion. Rather, it was torn apart by the G-forces being exerted on the shuttle from the outside.

The space shuttle launch system consisted of the orbiter itself; a giant, non-reuseable external liquid fuel tank; and two resueable solid fuel booster rockets. The shuttle was sold to Congress and the American people as a cheap, reuseable space launch system capable of more than 50 flights a year. The watch-word was "recycle". NASA originally wanted to make the main liquid fuel tank as well as the solid fuel external boosters recoverable and reuseable. But cost analyses showed that it was cheaper to abandon all three tanks in the sea during flight rather than recover them. Politically, NASA knew it couldn't get that past Congress. So NASA fudged the books, and told Congress that at least the external solid-fuel boosters could be reused.

Because of its design, the space shuttle was very, very, verrrry heavy. Everyone at NASA knew by 1975 that the shuttle wasn't going to be an effective space launch vehicle. The shuttle's payload bay was too small by half. And even if the payload bay were reconfigured and re-engineered to accept larger payloads, the shuttle was so under-powered it couldn't lift them. NASA would have been better off spending the shuttle's $7.5 billion development costs on building Saturn V rockets (for heavy lift payloads) and improved Delta V rockets (for smaller payloads), and building a space station to keep manned space flight going. Cost, too, was a problem: The total price tag for the shuttle program was more than twice the $90 billion NASA originally calculated. America spent more on the space shuttle than the cost of going to the moon, creating the atom bomb, and digging the Panama Canal COMBINED.

And let's not even talk about the shuttle's lack of escape system. Or that by placing the fuel tank on one side, it exposed it to massive amounts of debris during launch. Or its miniscule computing capacity. Or the way it exposed its heat shield to damage during launch.

The key to the Space Shuttle Challenger disaster was the solid rocket boosters (SRBs). Each SRB had six sections. Three of the sections were joined at the factory ("factory joints"), while the remaining three pieces were assembled at the Vehicle Assembly Building at Kennedy Space Center ("field joints"). The factory joints had asbestos-silica insulation applied over the joint. But the field joints were different.....

More about the Challenger disaster behind this cut.....

Morton Thiokol, a major space industry contractor, designed and built the SRBs. The field joints had to be flexible, because otherwise the intense internal pressures created by the hot expanding gases would blow the rocket apart like an overfull balloon. But Thiokol engineers knew that when the SRB was running, these field joints flexed outward due to these pressures. When defomed like that, the joints spurted flame. To seal the joints, Thiokol engineers put a rubber ring (an "O-ring") into the joint. The idea was that the pressure coming against the ring would flatten it, sealing the joint.

The SRB O-rings were designated a "Criticality 1" component. This meant that they had to function perfectly, all the time. If they failed, the orbiter and its crew would die. NASA rules said that there could be no backup to a "Criticality 1" component. Redundancy, however, was all right. The backup is there to provide redundancy in case of unforeseen failure, not to replace the primary device (leaving no backup.)

To provide that redundancy, a second O-ring was installed in the field joint. It, too, was designed to be a "Criticality 1" component.

But there was a problem with the O-rings. First, tests in 1977 showed that the field joints could deform so much that the first O-ring could be dislodged! This meant that for a few critical milliseconds as the rocket ignited (the moment of highest pressure), the gas could bypass the unflattened second O-ring and spurt out of the field joint. But NASA and Thiokol engineers believed that the length of time of joint deformation was so small and the amount of deformation so small that the risk was worth taking. Thiokol engineers also argued that since the test had been done in a horizontal (not vertical, or take-off) position, the second O-ring had not performed as well as it might have. Correcting for the effects of gravity on the horizontal O-ring, Thiokol asserted (without evidence), the second O-ring would seal. It was a terrible violation of NASA's "Criticality 1" requirement.

There was another problem, too. The O-rings were designed to resist heat, not cold. Thiokol engineers knew that if the O-rings were colder than 53 degrees Fahrenheit (12 C), the O-ring would not be flexible enough. It would not flatten, and it would not seal the field joint properly. Indeed, at very low temperatures (near freezing), the O-rings became so brittle that they shattered under pressure and failed completely! Marshall Space Flight Center managers in Huntsville, Alabama, discovered the problem with cold and the O-rings in 1978. But the Marshall managers only passed the concerns along to their counterparts at Morton Thiokol, and did not pass them up the NASA chain of command. This was a flagrant violation of NASA regulations. (It has been suggested that Mashall managers were trying to save the Morton Thiokol contract from being cancelled, and trying to save the jobs of the people at Thiokol whom they worked and whom they had befriended.)

The shuttle underwent four test launches. After the second test launch in November 1981, Thiokol investigators discovered that one of the first O-rings at the bottom of the rocket had suffered severe heat damage. Thiokol decided, however, that since none of the other O-rings in the SRB had been damaged, there must have been a flaw in the heat-resistant paste applied to the inside of the field joint. They reformulated the paste, and the next two flights came off without any damage. On the shuttle's second operational flight in April 1983, the paste failed again. But it did not fail on the next two flights, and both NASA and Thiokol engineers assumed there was no problem with the O-ring design. In 1984, O-rings were again found to be damaged. Thiokol now believed it knew why: NASA was using an air pressure test during field assembly to ensure that the O-rings were seated properly and not deformed. Thiokol asserted (again, without any evidence) that these air pressure tests were harming the paste by filling it with tiny air bubbles. At least the paste worked long enough so that the O-rings did not actually fail; they were just heavily damaged. Because air pressure tests could not be avoided, Thiokol and NASA in March 1983 reclassified O-ring damage to be an acceptable and normal part of shuttle operation.

Essentially, NASA and Thiokol were making a trade-off: They felt that ensuring that the O-rings were seated properly was more important than not damaging the heat-resistant paste. They felt that since the O-rings were performing "okay" in the launches so far, it was more important to risk damaging the paste that ensuring that both components worked well or redesigning the joint altogether (which would have cost Thiokol and NASA millions of dollars).

O-ring damage continued to be seen in subsequent launches, but none of the O-rings was compromised much. Nonetheless, engineers at Thiokol became increasingly concerned. In the summer of 1985, engineer Roger Boisjoly urgently requested that Thiokol and the Marshall Space Flight Center convene a joint team to develop a solution to the O-ring problem. He said the joint problem was so severe that a shuttle could be destroyed. NASA officials understood that fixing the O-ring problem would be a hugely expensive. But to their credit, they also realized they were flirting with disaster. So in the summer and fall of 1985, NASA began a process that would design latches to lock the O-rings in place (rather than use air pressure tests to determine if they were in place). In August 1985, NASA officials received a highly detailed briefing on the field joint problem, but concluded the shuttle should continue to fly. Thiokol officials set up an O-ring task force -- but then under-funded it and dragged out its work: Their contract with NASA was up and they didn't have the money yet to solve the O-ring problem.

Amazingly, in late 1985, Thiokol officials asked NASA to consider the O-ring problem solved. On January 23, 1986 -- just five days before the Challenger disaster -- NASA agreed to close the issue. It's not clear why either organization did this. Perhaps it was a political ploy to winnow the list of shuttle problems down, so that the second contract could be signed. (The problems would then go back on the list once the money was flowing again, and be fixed later.) Perhaps NASA wanted to lie to Congress, and show that it was making progress on shuttle issues. Perhaps NASA did not want shuttle problems exposed. Or, and most horribly, perhaps NASA and Thiokol managers (not engineers) decided without any evidence that the problem really wasn't important.

But it seems that, for NASA, the political risks of grounding the shuttle fleet and incurring the huge cost of fixing the field joints (years after it had certified the fleet as spaceworthy) were simply too high. For Thiokol, re-engineering the SRBs would eat into profits and jeopardize the renewal of its contract.

Essentially, NASA and Thiokol were playing Russian roulette. They knew that the risk of failure was high, and that the cost of failure was gigantic (explosion, death). But since they beat the odds once, then twice, they then made the decision that .... well, gee, maybe the risks weren't as high as originally thought. So NASA and Morton Thiokol reduced their assessment of that risk.

Look at a dice. There's a one in six chance of rolling a six. But just because you roll the dice six times, and no six comes up, doesn't alter the chance of rolling a six next time. Past performance is not an indicator of the level of risk. But both NASA and Morton Thiokol suddenly asserted it was.

* * * * * * *

Challenger was supposed to have launched on December 23, 1985. But round after round of bad weather created a month-long delay. NASA had a private satellite sitting in the shuttle's launch bay, and the satellite's owner was chomping at the bit to get the damn thing in orbit. Furthermore, NASA had two scientific missions due to launch that year. Both had to launch on time, or they would fail to reach Jupiter and slingshot around it to reach their appropriate destinations. A third scientific mission also had to launch on time, too, or it would miss its intended target: a probe designed to observe Halley's Comet. Challenger had to launch on time, land, and get refurbished in order to get these missions off the ground.

There were political reasons to rush a launch as well. America's first "teacher in space," Christa McAuliffe, was due to ride on Challenger. The news media tended to view NASA's failure to launch as a quality-control issue ("NASA fucks up yet again"), and a sign of over-caution. With intense media interest in the flight, NASA officials wished to avoid this kind of press criticism. Furthermore, President Ronald Reagan was due to give his State of the Union address the day after the launch. NASA officials hoped that Reagan would mention the shuttle launch in his speech, giving them a political boost in Congress.

The forecast for January 28 predicted temperatures close to 31 F (−1 C) at Cape Canaveral. The night before, NASA and Morton Thiokol managers and engineers held their usual "launch decision" telephone conference call. Thiokol engineers raised the issue of the O-rings not performing well at low temperatures, and said that NASA should not launch until the temperature was at least 53 F (12 C). Solid rocket booster project manager Lawrence Mulloy lashed out at Thiokol staff. "My God, Thiokol, when do you want me to launch, next April?" Other Marshall Space Flight Center staff also shouted and cursed at Thiokol managers. Mulloy attacked Thiokol's data and Thiokol's interpretation of the data, and accused Morton Thiokol engineers of trying to establish "new launch criteria" on the spur of the moment. Other Marshall staff characterized Thiokol's advice as "appalling". NASA officials said that even if the first O-ring failed, the second would seal (an assertion backed up by absolutely no test data).

NASA admitted it would not launch without Thiokol's approval. But this put the onus for not launching on Morton Thiokol. The Thiokol people on the conference call asked for a five-minute private caucus. As soon as the break began, the Thiokol managers began to huddle by themselves, without the engineers. Boisjoly and the other Thiokol engineers tried to intervene in this huddle, and reiterated their advice that it was unsafe to launch below 53 F (12 C). One of the Thiokol managers whipped around and sneered, "Why don't you take off your engineering hat and put on your management hat?" It was clear that the Morton Thiokol managers were making a political, not scientific, decision to launch.

When the conference call resumed, Thiokol managers (but not the engineers) reversed their decision and authorized a launch.

The Marshall managers did not raise Thiokol's concerns with their superiors. It's not clear that it would have made any difference had they done so, but they were supposed to do so -- and did not.

What happened on the January 27 evening conference call was a complete reversal of NASA's traditional approach to space flight. Usually, the contractor was required to prove that the spacecraft was safe to fly. This was a conservative approach that helped ensure a "fail-safe" culture of safety. But on January 27, NASA was demanding that Thiokol prove beyond a shadow of a doubt that the shuttle was unsafe to fly -- a complete reversal of its usual stand.

* * * * * * *

Ignition of the Space Shuttle Challenger's main engines occurred at at 11:38:00.010 A.M. Eastern Standard Time.

Just 0.678 seconds into the flight (T+0.678), launch pad cameras showed big puffs of black smoke coming from the starboard SRB near the aft strut that attaches the rocket to the main external fuel tank. Another smoke puff occurred at T+2.733, and a third at T+3.375. This smoke was caused by the opening and closing of the field joint in this area. Gases as hot as 5,000 degrees F (2,760 C) were spurting out of the rocket.

The amazing thing is that this had happened numerous times before. But the O-rings had eventually sealed, and both Thiokol and NASA had redefined the jets of hot gas not as failures of the rocket safety system but rather as "normal performance."

But this time was different. The cold weather meant that the O-rings were so brittle that they could not flatten out into place and form a seal. Furthermore, the second O-ring had shifted out of place due to rotation in rocket's joint. It could no longer form a seal, even if it peformed correctly.

The hot gas blew out the heat-resistant paste and bypassed both O-rings. The hot gas also severely damaged the O-rings. More than 70 percent of the first and second O-ring in the aft starboard booster was destroyed by this hot gas.

As the vehicle cleared the tower, the joints temporarily sealed as partially burnt fuel jammed into the field joints. But this blockage could last only seconds...

The shuttle's main engines achieved liftoff at 104 percent of their rated maximum thrust. At T+28, the engines throttled back so the shuttle wouldn't tear itself apart in the denser lower atmosphere. At T+35.379, the main engines reached their lowest thrust of 65 percent. Five seconds later, Challenger reached Mach 1. At T+37, the shuttle experienced a series of wind blasts that were the strongest ever recorded in the shuttle program. These winds probably caused some additional joint rotation, and dislodged the debris temporarily sealing the field joints. The gusts continued to hit the shuttle for the next 27 seconds (or until the shuttle blew up).

At T+51.860, Challenger passed into the less-dense outer atmosphere, and the main engines began throttling back up to 104 percent.

At T+58.788, the aft field joint in the starboard SRB gave way. A visible plume of flame spurted out of the joint and began destroying the strut attaching the rocket to the main external tank. Within one second, the rocket began to lose pressure as flame spurted out the field joint rather than the nozzles. At T+60.238, this flame was so powerful that it was now blasting away at the main external fuel tank.

At T+64.660, the flame changed shape. It had burned a hole in the main external fuel tank, and liquid hydrogen began blasting out of the tank. The hydrogen was ignited by the SRB flame, and acted like a new rocket source. The shuttle's main engines pivoted automatically under computer control to compensate. Pressure in the main external fuel tank began to drop at T+66.764. (It is unclear if NASA flight controllers noticed this. If they did, did they see it as a problem?)

At T+68, flight control told the Challenger crew that they were "go at throttle up." Flight control was telling the shuttle that as soon as their engines reached 104 percent of maximum rated thrust again, the shuttle was a "go" for orbit. Challenger's commander, Dick Scobee, replied, "Roger, go at throttle up."

At T+72.284, the aft strut attaching the starboard SRB to the main external fuel tank gave way. The rocket began to tear away from the rest of the shuttle. At T+72.525, the shuttle's crew felt a massive pull to the right. The last words the cockpit voice recorder captured came a half-second later, when Challenger's pilot, Michael Smith, said, "Uh oh."

At T+73.124, the main external fuel tank failed. It is not clear whether the fire had reached the interior of the tank and ignited the hydrogen stored there, or whether the loss of pressure in the hydrogen tank caused the internal barrier separating the liquid hydrogen from liquid oxygen to collapse. But whatever the cause, a breach occurred, the hydrogen mixed with the oxygen, and the hydrogen exploded. At the same time, the starboard SRB (still partially attached to the main external tank), careened into and struck the main tank.

At T+73.162, at an altitude of 48,000 feet (14.6 km), Challenger came loose from the main external tank. The air moving around the launch vehicle was so powerful, Challenger was torn apart. To those on the ground, it looked as if the shuttle had been destroyed by the explosion. But in fact, the shuttle was well away from the explosion and instead was torn apart by wind that exterted up to 20 times the force of gravity (Gs) against the spacecraft (well over its 5 G design limit). The explosion merely hurled the pieces of the shuttle away from the main explosion.

The two SRBs separated from the main tank and continued in uncontrolled flight for another 37 seconds. (NASA officials blew them up remotely.)

The crew cabin survived the breakup of the shuttle intact. Film of the disaster clearly shows the crew cabin flying clear of the explosion.

The crew cabin exited the explosion gases at T+75.237. The cabin continued to soar into the atmosphere until, 25 seconds later, it reached a maximum height of 65,000 feet (19.8 km). The crew suffered through 12 and 20 Gs of force for about two seconds, but the loads on the cabin fell to about 4 Gs after that. Within 10 seconds, the cabin was in free fall. These G forces would not have killed anyone in the cabin.

At least some of the astronauts were alive and conscious after Challenger was torn apart. Each astronaut had a Personal Egress Air Pack (PEAP), a briefcase-sized unit that supplied oxygen to the face. This was not a pressurized system, meaning that each person had to suck in oxygen from the PEAP; oxygen did not automatically flow from the PEAP without a person trying to draw air. Four of the seven PEAP units were recovered. Each astronaut had their own PEAP with a distinctive serial number. Astronaut Judith Resnik, mission specialist Ellison Onizuka, and pilot Michael Smith all had their PEAPs activated. It is most likely that either Resnik or Onizuka reached forward and turned Smith's on.

From breakup to impact took 2 minutes and 45 seconds. Resnik, Onizuka, and Smith used up air in their PEAP the entire time. Investigators could never be certain that the glass in the crew cabin remained intact after the explosion (because all of it was shattered upon impact). Had the cabin remained pressurized, the crew may have remained conscious. Had it depressurized, the crew could have survived but would have been unconscious.

Investigators also discovered that pilot Mike Smith attempted to regain electrical power in the cockpit after the shuttle ripped itself apart. Several switches on his electrical supply panel had been moved into new positions. These switches cannot be moved by unless they are pulled outward, moved into a new position, and then reinserted. Force alone (such as the G forces of the explosion or the impact with the ocean) cannot move them. It is clear that Smith attempted to restore electrical power to the cockpit, not realizing what had happened to the shuttle.

The crew cabin hit the ocean at 207 miles per hour (333 km/h). Each of the seven astronauts had remained in their chairs until impact. The cabin hit the water on its left side at 200 Gs. The force was so powerful that the left struts of the cabin -- which had survived the explosion undamaged -- were bent inward at a 90-degree angle. Each astronaut died instantly on impact.

On March 7, divers from the USS Preserver found the crew compartment on the ocean floor about 400 yards northeast of the Hetzel Shoals buoy (about 15 miles northeast of Cape Canaveral). The water at this point was about 100 feet deep. The remains of all seven crew members were found inside the crew cabin the next day. When the cabin was raised to the surface, the body of mission specialist Greg Jarvis fell out of the crew cabin, bobbed on the surface for a few seconds, and then slid back into the ocean. On April 15, the last day of wreckage recovery, Jarvis' body was rediscovered and returned to shore.

On July 28, 1986, Rear Admiral Richard H. Truly released a report which concluded:
  • The cause of death of the Challenger astronauts cannot be positively determined;
  • The G forces during the orbiter's breakup were probably not sufficient to cause death or serious injury; and
  • The crew may or may not have lost consciousness three or four seconds after breakup due to cabin depressurization.
NASA's lead investigator, Robert Overmyer, believes that most if not all of the crew were alive and conscious during the descent.

The bodies had been in warm water for nearly six weeks, and were highly decomposed. At the time, DNA testing had not yet been invented. Identifiable remains were turned over to the families on April 29, 1986. Dick Scobee and Michael Smith were buried at Arlington National Cemetery. Ellison Onizuka was buried at the National Memorial Cemetery of the Pacific in Honolulu, Hawaii. Greg Jarvis was cremated, and his ashes spread in the Pacific Ocean. Judith Resnik was also cremated, but her family refuses to say what was done with her remains. Ronald McNair was buried in his home town of Lake City, South Carolina. Christa McAuliffe was buried in Concord, New Hampshire, where she lived. But some portions of remains could not be identified. These remains were cremated, and buried beneath the Space Shuttle Challenger Memorial in Arlington National Cemetery on May 20, 1986.

* * * * * * *

President Ronald Reagan postponed his State of the Union address and instead gave a televised speech to the nation the day after the tragedy. Reagan also established the Presidential Commission on the Space Shuttle Challenger Accident. It became known as the Rogers Commission after its chairman, former Secretary of State William P. Rogers. Its Vice Chairman was Neil Armstrong, and other members included attorney David Acheson, aeronautics specialist Eugene Covert, Nobel Prize winning physicist Richard Feynman, aeronautics expert Robert Hotz, Air Force general and rocket expert Donald Kutyna, astronaut Sally Ride, aerospace engineer Robert Rummel, aeronautics engineer Joseph Sutter, physicist Arthur Walker, aviation executive Albert Wheelon, Air Force test pilot Chuck Yeager, and aerospace engineer Alton Keel.

The commission concluded that the failure of the O-rings was due not only to a faulty design, but also to due to the extremely low temperatures at Cape Canaveral on the day of the launch. The commission harshly criticized NASA and Morton Thiokol for failing to respond to the danger posed by the O-rings, for redefining the problem as acceptable risk, and for using a deeply flawed launch decision-making process. The commission said that Thiokol had presented incomplete and sometimes misleading information, management had overruled engineering judgment, and NASA management bypassed their own internal safety procedures.

Physicist Richard Feynman later famously immersed an O-ring in a glass of ice water on national television and dropped it on a table -- shattering it. It was a horrifying example of the kind of risks NASA had blithely taken. Feynman was so critical of NASA's organizational culture that he demanded the right to insert his own views about the disaster in an appendix to the Rogers Commission report. Feynman argued NASA's estimates of reliability were wildly unrealistic, often differing as much as a thousandfold from the estimates made by engineers.

* * * * * * *

The Space Shuttle Challenger Memorial was dedicated on March 21, 1987, at Arlington National Cemetery by Vice President George H.W. Bush. A rectangular bronze plaque is set in a vertical slab of gray Vermont marble. The plaque features a a seven-pointed star (representing the seven astronauts), with a bas-relief image of the shuttle at the center of the star and the likenesses of the crew at the points of the star.

Congress approved the memorial in June 1986. It was designed by Sarah LeClerc of the Army Institute of Heraldry, and approved by the family members of the shuttle crew. Donald Borja of the Army Institute of Heraldry sculpted it. The memorial is in a group with the Iran Rescue Mission Memorial and the Space Shuttle Columbia Memorial, near the Canadian Cross Memorial just behind the Memorial Amphitheater. Dick Scobee and Michael Smith are buried just a few feet away.

No one at NASA or Morton Thiokol was blamed or fired for the disaster. Michael Smith's widow sued NASA and Mulloy personally for the decision to launch, but federal courts ruled that Congress had barred such lawsuits.

NASA and Morton Thiokol redesigned the field joints to include a third O-ring, and finally began using latches (rather than air pressure tests) to ensure the O-rings were properly in place. The joints worked perfectly after that.

No comments:

Post a Comment