Friday, January 26, 2024

America's Worst Week in Spaceflight - An Annual Repost

It's an oddity of US Space travel that every mission that ended in loss of crew and vehicle occurred during one calendar week, although those accidents span 36 years. That week is January 27th through February 1st; while the years run from 1967 through 2003.  

January 27, 1967 was the hellish demise of Apollo 1 and her crew, Gus Grissom, Roger Chaffee and Ed White, during a pad test, not a flight.  In that article, Ars Technica interviews key men associated with the mission.  In the intervening years, I've heard speculation that we never would have made it to the moon without something to shake out a bit of the NASA management idiocy, but that may just be people logically justifying their opinions.  Like this quote from Chris Kraft, one of the giants of NASA in the '60s. 

There was plenty of blame to go around—for North American [built the Apollo capsule - SiG], for flight control in Houston, for technicians at Cape Canaveral, for Washington DC and its political pressure on the schedule and its increasingly bureaucratic approach to spaceflight. The reality is that the spacecraft was not flyable. It had too many faults. Had the Apollo 1 fire not occurred, it’s likely that additional problems would have delayed the launch.

“Unless the fire had happened, I think it’s very doubtful that we would have ever landed on the Moon,” Kraft said. “And I know damned well we wouldn’t have gotten there during the 1960s. There were just too many things wrong. Too many management problems, too many people problems, and too many hardware problems across the whole program.” 

The next big disaster was January 28, - the next day on the calendar, but in 1986, 19 years later.  Space Shuttle Challenger was lost a mere 73 seconds into mission 51-L as a flaw in the starboard solid rocket booster allowed a secondary flame to burn through supports and cause the external tank to explode.  It was the kind of cold day that we haven't had here in some years.  It has been reported that it was between 20 and 26 around the area on the morning of the launch and ice had been reported on the launch tower as well as the external tank.  O-rings that were used to seal the segments of the stackable solid rocket boosters were too cold to seal.  Launch wasn't until nearly noon and it had warmed somewhat, but the shuttle had never been launched at temperatures below 40 before that mission.  Richard Feynman famously demonstrated that cold was likely the cause during the televised Rogers Commission meetings, dropping a section of O ring compressed by a C-clamp into his iced water to demonstrate that it had lost its resilience at that temperature.  The vehicle would have been colder than that iced water.    


As important and memorable as that moment was, engineers such as Roger Boisjoly of Morton Thiokol, the makers of the boosters, fought managers for at least the full day before the launch, with managers eventually overruling the engineers.  Feynman had been told about the cold temperature issues with the O-rings by several people, and local rumors were that he would go to some of the bars just outside the gates of the Kennedy Space Center and talk with workers about what they saw.  The simple example with the O-ring and glass of iced water was vivid and brought the issue home to millions. 

There's plenty of evidence that the crew of Challenger survived the explosion.  The crew cabin was specifically designed to be used as an escape pod, but after most of the design work, NASA decided to drop the other requirements to save weight.  The recovered cabin had clear evidence of activity: oxygen bottles being turned on, switches that require a few steps to activate being flipped.  It's doubtful they survived the impact with the ocean and some believe they passed out due to hypoxia before that.  We'll honestly never know.

Finally, at the end of this worst week, Shuttle Columbia, the oldest surviving shuttle flying as mission STS-107, broke up on re-entry 17 years later on February 1, 2003 scattering wreckage over the central southern tier of the country with most debris along the Texas/Louisiana line.  As details emerged about the flight, it turns out that Columbia and everyone on board had been sentenced to death at launch - they just didn't know it.  A chunk of foam had broken off the external tank during liftoff and hit the left wing's carbon composite leading edge, punching a hole in it.  There was no way a shuttle could reenter without exposing that wing to conditions that would destroy it.  They were either going to die on reentry or sit up there and run out of food, water and air.  During reentry, hot plasma worked its way into that hole, through the structure of the wing, burning through piece after piece, sensor after sensor, until the wing tore off the shuttle and tore the vehicle apart.  Local lore on this one is that the original foam recipe was changed due to environmental regulations, causing them to switch to a foam that didn't adhere to the tank or stand up to abuse as well.  

In 2014, Ars Technica did a deep dive article on possible ways that Columbia's crew could have been saved.  They republished that on February 1, 2023, the anniversary of the disaster.  It's interesting speculation, very detailed, compiled by a man who claims to have been a junior system administrator for Boeing in Houston, working in Mission Control that day.  

Like many of you, I remember them all.  I was a 13 year-old kid midway through 7th grade in Miami when Apollo 1 burned.  By the time of Challenger, I was a 32 year old working on commercial satellite TV receivers here near the KSC and watched Challenger live via the satellite TV, instead of going outside to watch it as I always did.  Mrs. Graybeard had just begun working on the unmanned side on the Cape, next door to the facility that refurbished the Shuttles SRBs between flights, and was outside watching the launch.  Columbia happened when it was feeling routine again.  Mom had fallen and was in the hospital; we were preparing to go down to South Florida to visit and I was watching the TV waiting to hear the double sonic booms shake the house as they always did. They never came.

The failure reports and investigations of all three of these disasters center on the same things: the problems with NASA's way of doing things.  They tended to rely on "well, it worked last time" when dealing with dangerous situations, or leaned too much toward, "schedule is king" all as a way of gambling that someone else would be the one blamed for delaying a mission.  Spaceflight is inherently very risky, so some risk taking is inevitable, but NASA had taken stupid risks too often.  People playing Russian Roulette can say, "well, it worked last time," but having worked doesn't change the odds of losing.

 

 

9 comments:

  1. My dad knew most of the Mercury, Gemini and Apollo astronauts from working on Range Tracking and Instrumentation. Dad was a very mellow man, overall. But mention conspiracy theories that the Apollo 1 crew was still alive or that we didn't go to the Moon, well, deep deep anger.

    The measures taken after Apollo 1, from getting rid of a known issue with the wiring (so bad that it was banned on Navy fighters...) to tumbling the completed capsule to find all the garbage and loose things left lying around, to changing the air mixture and fixing the escape hatch to actually, you know, allow people to escape, all were things that made the Moon landings possible.

    Getting rid of the complacency that had set in by mid Gemini helped, too.

    As to Challenger, it still boggles me that a vehicle designed to launch year round from both the Cape, which was known to have periodic freezing temps during some winters, and Vandenberg, which also was known to have freezing temperatures during some winters (how do I know? Snow on Christmas Morn in 1973 at Satellite Beach, FL, just 20 miles south of the Cape, and lots of days riding by people's frozen yards (and the death of Florida's northern Citrus crops due to freezing in the mid 70's.) And 3 years at Santa Maria and 3 years at Vandenberg in the mid to late 60's, lots of incidents of frost and ice and very cold temps, no snow though...)

    As to Columbia, foam strikes were a known thing. Previous shuttles returned with non-critical areas devoid of heat tiles. And repair kits for both the white and black tiles existed, but were considered too expensive to actually stock and carry, and no training as to their use was ever done past testing the equipment.

    Regarding the exterior foam, yeah, had a NASA engineer tell me that due to changing the formula. The old used freon as part of the formula, and was strong and self-supporting, while the newer 'green' non-freon was fragile as all heck and heavier, which meant that they couldn't paint it with a flexible paint like on the old foam, and, hey, guess what the foam needed to make it stay on the tank? Oh, that's right, a flexible paintable external barrier...

    All of these disasters were, in hindsight, avoidable.

    And the Shuttle? One of the things that sold it to Congress and the American People was that there was always going to be another Shuttle on Pad Ready just in case. And there was also supposed to be an Atlas-mounted min-shuttle... And inflatable rescue balls/capsules/escape systems.

    Even worse, I fully expect some Anon Y. Mouse to come on and spout more conspiracy theories as to how none of these happened...

    ReplyDelete
  2. Working to reduce risk is (IMHO) a tricky thing. Initial intelligent risk mitigation steps are very valuable, but, like most things, the more you reduce risk, the less mitigation additional steps have. (Risk management needs to take into account overall performance objectives including technical, cost, and schedule).

    The tricky part: my experience indicates that there comes a point where additional risk mitigation steps are counter productive, and end up increasing the chances of an accident. (In part due to the fact that more adding more complexity to reduce risk can increase risk due to the additional complexity).

    This is complicated by (what I observe) our society's increasing risk aversion over that past half century.

    RE: Beans and conspiracy theories - I am amused by "we don't have the technology to go to the moon today, so how could we have gone 55 years ago..." I know we have better technology than we had back then. But NASA today is even more dysfunctional/bureaucratic/inefficient than they were prior to the Apollo 1 fire and society has moved on to other priorities...
    ART

    ReplyDelete
    Replies
    1. We have the technology. We just, as a nation, as a world, refuse to use the technology, refuse to push to do the difficult things.

      And government just gets in the way. How much further would SpaceX be with Starship if the various bureaus and departments weren't in the way?

      Sure there'd be more test articles littering the seas and oceans, but we'd be at least in orbit by now, if not on the Moon.

      Delete
    2. "Working to reduce risk is (IMHO) a tricky thing. Initial intelligent risk mitigation steps are very valuable, but, like most things, the more you reduce risk, the less mitigation additional steps have."

      I think of that as Heinlein's TANSTAAFL - There Ain't No Such Thing As A Free Lunch. At some point, risk has been reduced as much as it's possible to achieve and the only thing it can do is raise the cost.

      Zero is an asymptote. You can approach it but never reach it.

      Then there's the implicit idea that the people doing the assessment know every possible risk and that there's no such thing as the "one in a Godzillion" risk no one has thought of affecting the launch. "Who would have thought a meteorite would come at this exact moment and take out the vehicle, let alone the entire launch complex?" Or a million other things.

      "This is complicated by (what I observe) our society's increasing risk aversion over that past half century. " Which is a more polite way of saying it than blaming the lawyers, like I do.

      An analogy from my working life in designing hi-reliability electronics. Reliability is never improved by adding parts. You can add parts to reduced the chance of some failures, but "parts is parts" and the ones you add can fail, too.

      Delete
  3. For Challenger, NASA management specifically asked Thiokol management whether the vehicle met the procurement specification, which required the capability to launch from Vandenberg. Thiokol management told them it did. So they launched.
    For Columbia, the watermelons had forced NASA to change the foam used on the tank to be "green."

    ReplyDelete
  4. Just look at a picture of Columbia during its first launch and compare it to a picture of its last launch. The foam on the external tank changed color. The different foam came with a condition of a minimum launch temperature. Young children could have made a better decision than NASA did in 1986 with Challenger. When the person to blame could not be found, I lost any respect I may have once had for NASA.

    ReplyDelete
    Replies
    1. "Just look at a picture of Columbia during its first launch and compare it to a picture of its last launch. "

      Do you mean when they stopped painting the foam white to save weight?

      Delete
    2. They stopped painting the foam because the new foam was heavier and they had to save weight somehow.

      Delete
  5. The actual conspiracy is the selling and implementing of the factually false idea that the consensus reinforcement and risk averseness from tax-funded central planning produces more results faster than not doing central planning. Monopolies produce poor service and high prices. There is no exception for space exploration. How much further would SpaceX be with Starship if 100 million conservatives weren't willing to drop a dime on anybody who ignored the EPA and FAA? ICBMs are one of the biggest weapons, and NASA is gun control.

    ReplyDelete