Transterrestrial Musings  


Amazon Honor System Click Here to Pay

Space
Alan Boyle (MSNBC)
Space Politics (Jeff Foust)
Space Transport News (Clark Lindsey)
NASA Watch
NASA Space Flight
Hobby Space
A Voyage To Arcturus (Jay Manifold)
Dispatches From The Final Frontier (Michael Belfiore)
Personal Spaceflight (Jeff Foust)
Mars Blog
The Flame Trench (Florida Today)
Space Cynic
Rocket Forge (Michael Mealing)
COTS Watch (Michael Mealing)
Curmudgeon's Corner (Mark Whittington)
Selenian Boondocks
Tales of the Heliosphere
Out Of The Cradle
Space For Commerce (Brian Dunbar)
True Anomaly
Kevin Parkin
The Speculist (Phil Bowermaster)
Spacecraft (Chris Hall)
Space Pragmatism (Dan Schrimpsher)
Eternal Golden Braid (Fred Kiesche)
Carried Away (Dan Schmelzer)
Laughing Wolf (C. Blake Powers)
Chair Force Engineer (Air Force Procurement)
Spacearium
Saturn Follies
JesusPhreaks (Scott Bell)
Journoblogs
The Ombudsgod
Cut On The Bias (Susanna Cornett)
Joanne Jacobs


Site designed by


Powered by
Movable Type
Biting Commentary about Infinity, and Beyond!

« Grounded Fleet | Main | It's A Twofer! »

Complex Failure Bleg

One of the things that I'm working on is a series of case studies for failure of complex technological systems, particularly where a failure cascades (perhaps inevitably) into others. Columbia is a good example, in which the fragile leading-edge TPS was damaged during launch, which resulted in initial burnthrough during entry, which caused more internal damage, which resulted in a bigger hole in the wing, which resulted in increasing asymmetric forces on the vehicle, which resulted in eventual inability to keep the nose pointed forward, which resulted in the destructive breakup of the vehicle from aerodynamic forces.

Is anyone aware of similar cases (preferably non-space, e.g., the Bonefish fire)?

Posted by Rand Simberg at July 29, 2005 04:53 AM
TrackBack URL for this entry:
http://www.transterrestrial.com/mt-diagnostics.cgi/4086

Listed below are links to weblogs that reference this post from Transterrestrial Musings.
Comments

I remember seeing a television report about it (narrator was Jack Davenport, the series used to air on National Geographic too I think), from which I seem to remember that the chain was actually much more complex than Wikipedia shows.
For example, the "broken wheel" mentioned in the text had, if I remember correctly, undergone a checkup and/or a subsequent repair. A (relatively!) small strip of the wheel stripped off (molecularly), and shot through the cabin floor, where it remained stuck next to a passenger seat. There's no mention of those details in the Wiki text.

Best regards.

***
The Eschede train disaster was the worst train accident in German history. It happened on 3 June 1998, near the village of Eschede in the district of Celle, Lower Saxony.

The ICE high-speed train "Wilhelm Conrad Röntgen" was on the route from Munich to Hamburg. After stopping in Hanover at 10:30 am, the train continued its journey northwards. At 10:59 the train derailed. The cause of the disaster was a broken wheel. When passing a switch at over 200 km/h, the damaged wheel jumped off the rail. The locomotive was separated from the remaining train and the emergency brakes were activated; this braking had little effect as the braking distance was too long due to the velocity. The first four carriages made it through the road bridge that crossed the tracks at Eschede, but the fifth carriage drove against the pier. The bridge collapsed and buried two carriages, the rear carriages crashed into the wreckage and were totally torn apart.

While many passengers and the driver survived in the front part of the train, there was almost no chance of survival in the rear carriages. 101 people died in the disaster.

http://en.wikipedia.org/wiki/Eschede_train_disaster

***

Posted by Flex Flint at July 29, 2005 05:19 AM

Ah! Found the NG-link: http://www.nationalgeographic.com/channel/seconds/gallery_05.html

And two more links:
http://www.nationmaster.com/encyclopedia/Eschede-train-disaster
http://danger-ahead.railfan.net/features/eschede.htm#whatwrong

Posted by Flex Flint at July 29, 2005 05:34 AM

Books with multiple case studies of failures:

James Oberg, Uncovering Soviet Disasters

Matthys Levy and Mario Salvadori, Why Buildings Fall Down (all civil engineering, probably not as complex a cascade as you're looking for)

James R. Chiles, Inviting Disaster: Lessons From the Edge of Technology (start with this one)

Langeweische had a two-part article in the Atlantic on the airline crash caused by bad oxygen generators--cascading failures on the human rather than technical side--but it's subscriber only now.

Posted by Karl Gallagher at July 29, 2005 07:42 AM

Perhaps the Concorde disaster could qualify. Its still Aerospace related though.

On take off a tire is slit open by a peice of metal that fell of an aircraft that took off just prior. The tire disintergrates and a chunk of tire weighing several pounds thunks against the underside of the wing. Even though the peice of tire doesn't penetrate through the aircraft skin it produces a shockwave of energy that propogates throughout the fuel contained in the wing. When the shockwave reaches the leading edge of the wing it finds a way out through a service panel. This causes fuel to burst out through the service panel and spray out over the number 2 engine. The hot exhaust of the engine ignites the fuel spray and begins to melt through the aircraft skin and control surfaces of the wing. Eventually the wing degrades to the point where it can no longer control the aircraft.

Posted by Josh Reiter at July 29, 2005 08:37 AM

"Set Phasers on Stun" treats a collection of 20 or so engineering failures lightly as somewhat fictional stories told from the point of view of the people who were involved (either directly or afterwards). It provides real references.

Posted by Karl Hallowell at July 29, 2005 09:27 AM

The Bonefish fire isn't a particularly good example - it's a fairly straightforward casualty. (One interesting note: The seawater leak was the result of poor maintenance and preservation.)

A better is example is Thresher where what is believed to have been a minor casualty (a small leak) cascaded into something much greater because the leak disabled the propulsion system. The ultimate cause of the loss was poor design and construction of the very system supposed to save their lives - the Emergency Main Ballast Tank Blow system.

Posted by Derek L. at July 29, 2005 10:16 AM

Forgot to add; Another interesting book on technological accidents is:

Perrow, Charles; Normal Accidents.

Posted by Derek L. at July 29, 2005 10:18 AM

Check out the NTSB for aviation reports. Just about every aviation crash follows the pattern of some small thing early on creating a chain of events leading to a crash.

Posted by David Masten at July 29, 2005 10:55 AM

"Arching between cables?" Gimme a break. One good thing about Wikipedia is that annoying spelling or grammatical errors are easily corrected even if you aren't an expert on the subject matter.

I know a few people whose disdain for Wikipedia is particularly intense. I love it.

Posted by Obi-Wan at July 29, 2005 03:44 PM

Rand,

The ValuJet crash was a good example.

Contract Maintenance improperly labels O2 generators.
Generators are improperly packaged (without proper safeing).
More generators than allowed by shipping regulations were placed in one box, and one cargo compartment.
Package was mislabeld as to contents.
Box with O2 generators was placed in cargo hold, where one ignites.
Heat from generator causes others to ignite, causing cascade reaction that produces large amounts of heat and pure O2.
Box was stacked on top of a tire, which proved to be a handy fuel source.
Because the generators were producing their own O2, the cargo compartment (which was designed to limit available air to a fire to a specific volume) did not function as designed.
Eventuall the tire (which was pressurized) blew, possibly rupturing the cargo comaprtment lining.
Heat form the conflagration eventually melted floor beams, impinging on flight control cables, which led to an unsustainable aircraft attitude.

Posted by Steve at July 29, 2005 04:12 PM

Nearly all submarine mishaps involve a failure cascade with 'delta P' as the instigator or finale. Well-illustrated 'coffee table' book but sparce on technical details: LOST SUBS (From the Hunley to the Kursk) Spencer Dunmore 2002 ISBN 1-903985-48-X

Posted by SpaceCat at July 29, 2005 08:28 PM

I learned about the Kaprun funicular fire (http://en.wikipedia.org/wiki/Kaprun_disaster) from an episode of the National Geographic series mentioned above.

When the decades-old trains were overhauled their appearance was updated to modern sensibilities but the mechanical work was careless. A cheap household electric heater was installed in the driver's compartment. It was overused, overheated, and ignited leaking brake fluid. The fail-safe brakes locked when deprived of fluid and held the train in the tunnel as it became a furnace.

Posted by FC at July 30, 2005 03:44 AM

Three Mile Island may be a good example, but there were a few critical operator errors involved in the event (i.e.: the technical failures were not *solely* responsible for the accident):

- a valve in the secondary coolant loop shuts down (due to an error in maintenance work, IIRC);
- this triggers a shutdown in the feedwater pumps;
- auxiliary feedwater system kicked in at this point, but during earlier maintenance a valve isolating the auxiliary system had been closed and erroneously not reopened, preventing cooling water from getting to the steam generators;
- water remaining in the steam generators boiled off, halting heat transfer from the primary loop;
- rising pressure in the primary (core) loop then triggered an automatic shutdown of the reactor proper (but, like stopping a train, it takes a while after you put on the brakes);
- primary coolant pressure continued to rise despite declining heating in the shut-down reactor, since there was nowhere for the heat to go;
- a pressure relief valve on the primary loop was triggered, releasing the pressure;
- the valve then failed to close again once pressure dropped -- it was known to have a small leak and the temp sensor on the outlet pipe was known to read high, so the information that should have tipped off the operators that the valve hadn't closed was disregarded;
- with pressure now decreasing in the primary loop due to the stuck relief valve, the coolant there begins to boil off;
- however, the relevant sensors can't distinguish the difference between the steam and water, and give a false reading of a *high* coolant level;
- the emergency water injection pumps automatically (correctly) began pouring water into the primary loop to make up for the actual loss of coolant;
- the operators, seeing the coolant level as high and trained to prevent overfilling of the reactor (lest it rupture the primary piping), shut the auxiliary pumps off;
- seeing the pressure still falling in the primary loop, the operators shut down the primary loop pumps to head off potentially destructive cavitation;
- with the primary coolant boiling off and not being replaced, the core got hotter and hotter, until the fuel cladding failed and a portion of the fuel itself melted into a heap in the bottom of the reactor vessel.

Posted by T.L. James at July 30, 2005 10:00 AM

See generally H. Petroski's book entitled "Design Paradigms - Case Histories of Error and Judgment in Engineering" (Cambridge 1994) for discussion of, among others, Apollo Pedestal, Kansas City Regency, Dee Bridge, Tacoma Narrows and other instances which seem to present facially structurally simple design issues, perhaps even more pernicious than "complex" systems because of susceptibility to overlook subtle potential complex failure modes. Also, obviously, structural issues in Twin Towers. Also you may be familiar with USS Thresher, a nuclear sub which imploded off the Azores in the 80's when power plant failure combined with unforseen anomalies in plane and buoyancy tank design, as I think I recall from replicating the circumstances in sub trainers in Charleston in the 70's, caused loss of forward way and settling of stern, and ultimately a stern-down slide to crush depth. As an aero engineer who flew in P-3's in 1970's, I am also generally familiar with early failures of wing root structures and fatal crashes of Lockheed Electra passenger aircraft (civilian P-3 counterpart), and subsequent reduction of wing length and controlling propeller rotational speed to prevent harmonics which caused the wings to fail (so-called "whirl mode failure"). Perhaps most insidious failures aboard the P-3, however, involved failure of coffee pot or electric skillet back in crew galley, thus cutting into crew performance with consequent secondary, tertiary, quaternary cascading effects on antisubmarine mission effectiveness and consequent ultimate ire of debriefing officers on the ground (just kidding about this last).

Posted by Jim Ewing at July 31, 2005 01:01 PM


Post a comment
Name:


Email Address:


URL:


Comments: