Slashdot Mirror
Stupid Engineering Mistakes
lee1 writes "Wired has bestowed on us a list of the ten worst engineering mistakes of all time. We have the St. Francis Dam designed by 'self-taught' engineer William Mulholland, which burst and wiped out several towns near LA; the Kansas City Hyatt walkway collapse; the DC-10, and more, but my favorite is the one I'd never heard of: a giant tank of molasses that ruptured in 1919 and sent 'waves of molasses up to 15 feet high' through Boston, killing 21."
Built on national pride, it's become the world's largest albatross.
A feeling of having made the same mistake before: Deja Foobar
What on earth were they planning on doing with such a huge stockpile of molasses?!
Osaka built the world's first sports stadium with a movable roof, which malfunctioned shortly after inception, and the company that made it went bankrupt. The roof has been stuck for the past 5 years. Incidentally, the stadium was built on rubbery landfill, so whenever audiences jump up and down during rock concerts, it causes earthquakes in the neighborhood. Osaka also built a new airport on an artificial island that is sinking into the sea, so it may become the world's first underwater airport. Seoul has had various engineering disasters also, including a department store that collapsed and killed hundreds of wealthy housewives.
this disaster involved a couple morons on a drilling rig in a lake forgetting to carry the two, hitting a mineshaft, and draining the whole lake and part of the gulf of mexico into the mine, along with several ships, etc etc.
VLC FOR MAC IS DYING! IF YOU DEVELOP, PLEASE SAVE IT!!
"much less the weight of the spectators standing on it"."
The History Channel had some coverage on their Modern Marvels series I think of this incident. Besides what you mentioned, the most damning was those inspectors did something like a 10 minute inspection...for the whole hotel, walkway inclusive.
The inspectors didn't do their job. This is much less about blaming one person or body, but usually these disasters had a whole sequence of things ignored that in cumulative resulted in disaster.
Case in point was the St. Francis damn--the issue had squat to do with a person who was self-taught. It had to do with the community, other engineers, excavators/construction--all had opportunities or should have had opportunities to correct or identify problems, but they were overlooked, ignored, politically side-barred.
http://en.wikipedia.org/wiki/Lake_Peigneur/
c o.html/
Basically, an oil rig, drilling in the middle of the lake, punctured a mineshaft below the lake (mining for salt). The end result was the entire lake draining into the mine below it. Fortunately, nobody was hurt.
From: http://members.tripod.com/~earthdude1/texaco/texa
The water of Lake Peigneur slowly started to turn, eventually forming a giant whirlpool. A large crater developed in the bottom of the lake. It was like someone pulled the stopper out of the bottom of a giant bathtub.
The crater grew larger and larger (it would eventually reach sixty yards in diameter). The water went down the hole faster and faster. The lake had been connected by the Delcambre Canal to the Gulf of Mexico, some twelve miles away. The ever-emptying lake caused the canal to lower by 3.5 feet and to start flowing in reverse. A fifty foot waterfall (the highest ever to exist in the state) formed where the canal water emptied into the crater.
The whirlpool easily sucked up the $5 million Texaco drilling platform, a second drilling rig that was nearby, a tugboat, eleven barges from the canal, a barge loading dock, seventy acres of Jefferson Island and its botanical gardens, parts of greenhouses, a house trailer, trucks, tractors, a parking lot, tons of mud, trees, and who knows what else. A natural gas fire broke out where the Texaco well was being drilled. Let's not forget the estimated 1.5 billion gallons of water that seemed to magically drain down the hole (does the Coriolis effect come into play here?). Of course, there was the great threat of environmental and economical catastrophe.
In 1814 in in London town,
g ht=&
a flood of beer came to drown.
http://www.qi.com/talk/viewtopic.php?t=121&highli
Time flies like an arrow. Fruit flies like a banana.
I can't quite remember, but I seem to recall that the records are scanty on this point -- it may be that the designers of the ship just didn't have the expertise and understanding of buoyancy of later shipwrights, or it may be that there was some kind of kickbacks or other shenanigans that interfered with the building and compromised the design.
When I say "if you look at the ship," though, I am being literal -- because you can. The really interesting thing about the Vasa is that it sank not far from Stockholm harbor, in waters that had a unique mineral consistency. Unlike other parts of the world, for whatever reason the waters in this area were particularly unfavorable to the shipworm. Normally a wooden ship like the Vasa would be eaten up. The Vasa, however, was merely covered with silt at the bottom of the bay, where it lay for hundreds of years.
Eventually -- and again, memory fails me but I believe it was sometime around the 1970s -- the location of the Vasa was discovered and work began to bring it to the surface. Today the entire ship is on display in a museum in Stockholm. The museum building was actually built up around the ship itself. A lot of repair and preservation work had to be done, including plastination of the wood, but it is mostly intact except for the original painting. You can't go onboard, but you can walk around it and view the hull from all sides. It is literally the closest you'll ever get to a 17th century wood-hull sailing vessel -- about five meters away. They've also built a facsimile of the interior decks that you can walk through -- if walking is the word. (Let's just say they made people smaller in those days.)
The museum has salvaged all kinds of other goodies from the ship as well, from cannon to tools to even the bodies of some of the original sailors, all of which are on display. If you get the chance you should check it out -- if you're at all into things nautical, it's a one-of-a-kind experience.
Breakfast served all day!
This one isn't quite on topic, but it keeps with the mood... Lake Peigneur: The Swirling Vortex of Doom
When I think of engineering mistakes, the Cypress Freeway comes to mind. A double-decker freeway built on soil that isn't solid in an earthquake-prone area is a disaster waiting to happen.
The former double-decker section of 880 has since been replaced with a new, single decker structure a bit to the west of the original alignment. The cost of that new, short freeway section was $1.13 billion dollars, more expensive than the costs of LA's Century Freeway (105), IIRC.
I agree with a poster above that this shouldn't be listed under "funny" as all of those mistakes cost well over 1,000 people their lives, if I remember the article correctly. But it seemed to focus on the fact that people's lives were lost in just about all of those. I would have placed a number of other engineering mistakes in that list just because of the nature of the mistake.
For example, the bridge (the name of which I can't remember) from the early part of the 20th century that bent and twisted under high wind until it finally just fell apart. Loss of life? I don't believe so, but it was a spectacular destruction.
The Johnstown Flood, perhaps? A lot of people were killed in that flood, and it was caused by engineering of a sort. The dam itself seemed to be stable until a lot of critical components, such as iron rods, were replaced with such highly stable components as dirt and manure, at least according to various web sites and documentaries. Sure, that wasn't a fault of the original design, but the "remodeling" is most likely a very important factor that resulted in the deaths of over 2,200 people.
I found it particularly interesting that the article mentioned how something happened 200 years before Titanic then failed to mention the Titanic itself. Based on the documentaries I've seen, the bolts that were used to hold the steel plates together were cheaply made and severely weakened under the frigid water of the north Atlantic. That was an engineering/design flaw from the beginning.
New Orleans. Oh, yeah! Let's design and build a city with an ocean on one side and a lake on the other and - here's the clincher - we'll make it below sea level! Yeah, baby! Party on! Enough said.
Seriously. I don't know what criteria this person used for the "worst" engineering mistakes, but it's clear to me at least that he really doesn't know what the hell he's talking about.
The Overrated mod is for reversing inappropriate, positive mods, not for voicing disagreement with a post.
>>what seemed an innocuous change completely changed the dynamics of load bearing.
I studied _ART_ in college and I spotted the flaw a mile away.
The specs called for two "C" shaped beams to hug a metal rod as so - ]|[
They were assembled like this - [|]
You have _much_ more strength when all vertical peices are touching, relying on the compression strength of the steel. They were assembled more like a rod going through a box. Now you have your force on horizonal portions of the beam. A little bit of bending and BAM! no more walkway.
Like many engineering disaters, its not the plans that were wrong but the changes made to them. Personally, I found it amazing that the construction crew didn't see the flaw.
Ok, I'm going back to making pretty pictures...
Know what I like about atheists? I've yet to meet one that believes God is on their side.
And really, the humor section? I know being killed by a flood of molasses is novel, how is having a walkway full of people falling on your head funny?
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
Yes, and they even had back then the culture "Don't blow the whistle to management that the project is doomed".
:-)
I also visited the museum (quite impressive indeed) and there they told that they used to test ships for their stability by having a number of soldiers run from one side of the deck to the other in a coordinated fashion to see if the ship would start to sway. And sway it did, that strong that they had to stop the test to keep it from capsizing. But who wanted to tell the king that his wondership, the one he meant to dominate the Baltic Sea, was not even seaworthy for a pond ?
So everyone kept silent, the ship went under having hardly cleared the harbour, and the best: Afterwards noone could be hold responsible: The master shipbuilder having designed the ship had died before the launch, his successor only inherited the design at a very late stage and couldn't make any substantial changes, and the King, well... you don't hold the King accountable !
...but I thought "Bridges Are Easy"....
That was not the critical flaw. The original design and the implementation used box beams. The implementation failed because the lower levels were attached to the top bridge, not directly to the rod as designed, thereby increasing the weight that pulled on the joint between the top bridge and the rod. The WP article explains it quite nicely and has pictures.
Because the guy honestly didn't care.
s
He (him) fsck'd huge parts of the west out of their water rights to get an ROI out of his investments in L.A.
The damn breaking was terrible PR. I believe it only troubled him because of the fear he would be found liable for the damage.
http://en.wikipedia.org/wiki/California_Water_War
You can still find *giant* chunks of concrete in the flood basin in the east end of the san fernando valley. I was honestly surprised to find them there.
http://www.maxineudall.com/2010/02/should-economists-be-sued-for-malpractice.html
The molasses flood was not an engineering mistake. The basic design of the structure was ok, the disaster is believed to be most likely to been have caused by shoddy contruction techniques and/or overfilling plus pressure buildup due to fermentation of the molasses in the tank.
"I studied _ART_ in college and I spotted the flaw a mile away."
Yes it shows that you studied art and not engineering. We actually studied this failure in one of my classes. The poorly welded box beams probably contributed to the failure but the much larger flaw was changing the support from one in which the box beams would only be supporting the weight of one floor to one in which they would be supporting the weight of all the floors. As I recall a junior engineer approved the change without consulting with more experienced engineers. The construction crew is not at fault because they built the structure according to approved plans and field changes.
No discussion of engineering disasters is complete without mention of PEPCON. First, build a factory 10 miles from Las Vegas. Use it to manufacture ammonium perchlorate -- a component of rocket fuel. Store the stuff in aluminum containers. BTW, aluminum is the other component for the rocket fuel. Then start welding nearby. Oh, and make sure you put the factory on top of a gas main.
http://en.wikipedia.org/wiki/PEPCON_disaster
There's some great footage of it here:
http://www.apechild.com/videos/pepcon.mov
You'll never see a better demonstration of speed-of-sound vs speed-of-light. You see massive explosions and shockwaves (taking out trees and cars) several seconds before you hear them.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
Say what you will about the guy, but he came up from being a ditch digger to chief engineer of DWP, you don't see that kind of stuff anymore.
I grew up very near the St. Francis dam disaster, we used to hang out on old giant slabs of concrete that were miles downstream from the former reservoir.
"'Yrch!' said Legolas, falling into his own tongue."
The truly sad aspect of the R-101 disaster is not that it crashed, but that the crash utterly killed any chance that the R-101's sister airship, the R-100, would gain public acceptance.
The two ships were built simultaneously, to the same set of government design specifications. The R-101 was designed by government engineers with an effectively unlimited budget, and no penalties for failing to meet specifications. Because a government agency was building it, the press were treated to frequent and highly colored bulletins about the R-101.
The R-100 was designed by a private firm, under a strict budget, with limited access to design information about the R-101. It was built with much less publicity and launched with no fanfare at all.
The R-100 made a successful trans-Atlantic test flight, was several knots faster than the specification called for, was highly maneuverable, and had a considerable payload capacity. It performed almost flawlessly, and was fairly economical to operate. (The Wikipedia article makes a bit much of the R-100's problems, such as the tail cone collapse; the engineers decided that the tail cone was unnecessary.)
The R-101 was grossly oversized and overweight, poorly stressed, and had been lengthened by some yards at the eleventh hour. Because of pressure to outperform the R-100, it was sent on an intercontinental flight before its local flight tests (which would probably have revealed its weaknesses) were completed. When it crashed, it took with it any chance that the R-100 would be followed up, even though the R-100 was a nearly unqualified success (for a prototype, anyway).
Dig up a copy of Nevil Shute's Slide Rule for an entertaining and sometimes harrowing account of the two rival airships.
I figure by 2030 or so my 6-digit UID will be something to brag about.
An incident I particularly remember involving building design was back in the early 80's, in Canada I believe. The architect designed a large circular building (a convention center or hotel, I disremember which) with a domed roof. Somebody later decided the edge of the domed roof was a great place for a jogging track, without studying the wind patterns the roof created. After the building was opened, with its unplanned addition, several people were blown off the track to their deaths.
I figure by 2030 or so my 6-digit UID will be something to brag about.
Frank Black wrote not one but two songs about Mulholland: "Ole Mulholland," and "The St. Francis Dam Disaster." Apparently he made quite an impression on the guy. I didn't connect these two until I saw this article, by the way.
My book, podcast
You read it correct. Basically, the wing engine tore away, taking out the hydralics. Sadly, the crew never knew the full state of what was happening. Walt lux (a family friend and my father's co-worker) was a senior captain. He did exactly as he was trained to do. Sadly, without the indicators, they never stood a chance.
I have thought about the fact that my father had one in 61 chance of being the co-pilot on that craft. In fact, airlines pilots (back then, anyways) tend to fly together and would pick the same schedule. Walt and my father flew together a great deal at that time, so it was probably like 1 in 5 chance. Weird to contemplate.
I prefer the "u" in honour as it seems to be missing these days.
Vasa had a complement of 445, of whom it is not clear how many were lost. The HMS Royal George, however, sank just off Spithead on August 29, 1782 in very similar circumstances to the Vasa with the loss of eight hundred, including an admiral of the fleet. An inquest concluded that her loss was due to structural failure. This was one of the worst marittime disasters of all time, and I'm surprised that the loss of the Vasa, and not of the Royal George, is on the list.
The Tacoma Narrows bridge didn't fail due to resonance.
Read that first line again.
It was not resonance, your first year, second year, calculus, dynamics and control systems books all lied to you. Lied. Not truthful. Not correct.
Read: K. Billah and R. Scanlan, "Resonance, Tacoma Narrows Bridge Failure, and Undergraduate Physics, Textbooks;" American Journal of Physics, 1991.
It was not a time dependant thing, therefore, not resonance. The bridge was shaking NOWHERE near its resonant frequencies. The motion of the bridge actually induced "negative damping" . That would sort of be like pulling your parachute and having it drag you to the ground faster and faster as you gain speed. Sounds weird, but totally true. They show in that paper that the bridge under the wind loading becomes a self excited structure and, at a critical wind speed, the eigenvalues of the bridge stucture change sign, causing the bridge to enter an exponentially increasing vibrational state, eventually breaking the bridge down.
I built a cool model of the Tacoma narrows bridge, with controllable air flow, and reproduced this behavior for a college course in experimental design. It was neat to visually watch eigenvalues change in an experiment.
Oh the physics of pulling wool over eyes is so fun. BTW, that "doubling the loading that any physics student could understand" bit in the other posts. Right. Most physics students can't tell you if the box slips downhill or uphill using a free body diagram. Give me a break.
There was a time for a while in the US that everyone and their brother was afraid we'd run out of water, tomorrow (EVERYBODY PANIC). Manufacturers (temporarily) switched production, almost exclusively, to front-load machines to capitalize on that fear. It turns out that the only people afraid enough to actually use the damn things were the people who live in deserts (I'm looking at you southern California). They're the ones that keep foisting abominations like low-flow showerheads on those of us smart enough to live close enough to stable water supplies.
n.b., Just so you know I'm at least half-joking about SoCal, I live in New Orleans, where we occasionally have a little too much water.
I don't recall what the Engineers did about it. They probably just repointed the mortar, slapped on a fresh coat of paint and sneaked back to the barracks.
Check out http://en.wikipedia.org/wiki/Tay_Rail_Bridge. Even in 1879, they couldn't build railway bridges. The Tay Rail Bridge disaster was the reason for compulsory registration of civil engineers, and brought on a large degree of over-engineering in all civil engineering projects. Over-engineering meant that it wouldn't fall down, but it also meant it'd be vastly more expensive to build.
Grab.
Actually, if you look at those things, the real common theme was that they were designed or modified by people who _weren't_ real engineers. E.g.:
- a dam is built by a "self-taught engineer" who can't even get the foundations right
- a ship design is modified by a king who has no flippin' clue about ship design. He demanded changes like cutting extra portholes right above the water line, loading extra guns and other stuff, and so on. The final design was basically the king's, not the design of a real shipwright.
- a huge container for molasses is designed and its building supervised by a beancounter with _no_ engineering background whatsoever, and whose only concern was getting it built quickly and cheaply.
Etc. Sorry, you can't say "never trust the engineers" when, in fact, those mistakes were made by non-engineers.
Want another common theme? How about ignoring testing or warning signs that it's about to fall apart. E.g.:
- when the dam started to crack, the "self-taught engineer" just ignored it
- the molasses container was (A) never tested, e.g., by filling it with water, and (B) when a worker complained that it leaked heavily, the beancounter just covered the problem by having it painted brown.
- the Vasa, as other posters have noted, was in fact tested before being lanched, but noone had the courage to tell the king that his design doesn't work. In effect, again the warning signs existed, but were effectively ignored.
And the third thing is: don't think those are just historical trivia, because the exact same things happen nowadays with software. Everyone loves to spew the "colleges don't teach engineering" or "it's time programmers started acting like engineers", but some of the most catastrophic mistakes come from people who had _neither_ a CS or engineering college, _nor_ reasonable work experience or training to bring them up to par. I'm not even sneering (mainly) at the actual coders, because lot of those mistakes were from some manager or customer demanding/making some catastrophic change or imposing some impossible deadmark or policy. (Remember the Vasa and the king.)
E.g.,
- a financial institution restates its earnings by 1 _billion_ dollars, because some Excel spreadsheet programmed by a beancounter with _zero_ engineering or programming background... guess what? Mis-calculated by a whole billion dollars.
- a radiotherapy machine, using lead blocks to cover the parts of the patient that shouldn't be irradiated, had a problem using more than IIRC 4 lead blocks. So a doctor takes it upon himself to hack it to use non-rectangular blocks to the same end. The result: the program mis-calculates and some people are given a lethal dose of radiation.
And that's just the spectacular stuff. I'm sure almost everyone has their own stories where someone else's intervention had catastrophic results, even if in less spectacular ways.
A polar bear is a cartesian bear after a coordinate transform.
A single protective relay tripped in Ontario,
And it was my fault. Or perhaps my friend's. We'd gone over to his house after school (this was in Toronto) to watch TV. I turned on the TV and he went into the kitchen to plug in the kettle for some tea. While he was out there the image on the TV started to shrink and flicker as the power went flaky, so I called out to him "unplug the kettle, you're blowing a fuse".
If I'd only said that a few moments sooner...
-- Alastair