Mechanic's Mistake Trashes $244 Million Aircraft 428
Hugh Pickens writes "An accident report is finally out for the Air Force E-8C Joint Surveillance Targeting and Attack Radar System that had started refueling with a KC-135 on on March 13, 2009 when the crew heard a 'loud bang throughout the midsection of the aircraft.' Vapor and fuel started pouring out of the JSTARS from 'at least two holes in the left wing just inboard of the number two engine.' The pilot immediately brought the jet back to its base in Qatar where mechanics found the number two main fuel tank had been ruptured, 'causing extensive damage to the wing of the aircraft.' How extensive? 25 million dollars worth of extensive. What caused this potentially fatal and incredibly expensive accident to one of the United States' biggest spy planes? According to the USAF accident report, a contractor accidentally left a plug in one of the fuel tank's relief vents (PDF) during routine maintenance. 'The PDM subcontractor employed ineffective tool control measures,' reads the report. Tool control measures? 'You know, the absolutely basic practice of accounting for the exact location of every tool that is used to work on an airplane once that work is finished.' Air Force Chief of Staff Gen. Norton Schwartz just told Congress, 'there is a JSTARS platform that was damaged beyond economical repair that we will not repair.' So, if this is the one Schwartz is talking about, then one mechanic's mistake has damaged a $244 million aircraft beyond repair."
Shit Happens (Score:5, Insightful)
I've been an A&P for over 35 years and I've seen worse.
(by pilots and mechanics)
Only 244 million? (Score:2, Insightful)
244 million? Isn't that minuscule? CEOs regularly crash the stock market. But at least they take responsibility! Like... becoming CEO somewhere else?
Re:RFID (Score:4, Insightful)
"You can get nearly 2,000 tags for about $100"
You or I could, but the essential middlemen selling the same stuff to the government would add at least three zeros to the end of that figure
Re:I'm not really understanding... (Score:5, Insightful)
Re:Government Contractors (Score:5, Insightful)
In the real world, faced with $244,000,000 in lawsuits, the contractor folds up and declares bankruptcy.
Then everyone will have a laugh and the taxpayers will pick up the tab.
Re:Utter nonsense (Score:4, Insightful)
It is ass-covering of the lowest order to blame a lowly mechanic for what is obviously a design flaw. A simple sensor to monitor the presence of a plug
Terrible design mistake because now someone needs to maintain, replace, test, and probably F-up that sensor. Also its heavy. The better design involves multiple permanently installed frangible disks on extra vent piping.
See how hard design is? Finding incompetence is always easier than designing around it. First guess is usually wrong. That's probably what happened to the A+P mechanic, too.
Re:RFID (Score:5, Insightful)
The government is big on COTS hardware/software, and only turn to contractors for specialized circumstances. Those extra zeros come from the unusual design requirements and low volume orders.
Take the x thousand dollar hammer example. On the surface, that seems absurd, since one can buy a hammer for less than 10$. But when the hammer is going into space and is made of a difficult to machine titanium alloy (tool steel shatters at cold temperatures), is egonomic even through spacesuit gloves, is lightened without reducing mechanical efficiency (makes sense at an estimated 1000$/pound/launch), and only 10 are made (despite flat machining costs), that X or XX thousand dollar price tag seems very affordable.
The same thing happens in other areas. I work on submarines and some components use joysticks. Sure, commercial joysticks can be obtained for under 100$, but a waterproofed, pov only motion, high durability (sailors treat equipment like crap, and failure is not an option) piece of clockwork machinery that maybe 50 will be made, you are looking at just shy of XX thousand per.
Re:Affirmative Action (Score:4, Insightful)
Any alternative is better than Affirmative Action. Giving someone a job because they belong to a minority is equivalent to not giving someone a job because they aren't in the minority, which is racist/sexist.
Re:Utter nonsense (Score:3, Insightful)
Remember that the airframe for this airplane (a KC-135) is basically a late 50s-early 60s design/build.
Whatever. Back then engineers had to be smarter because they couldn't rely on computers. The days of iron men, not heavy iron mainframes... Age is no excuse for poor design, assuming thats what you meant.
More likely, since this has not been a popular failure mode over the past half century, the cost of designing it out probably exceeds the cost of just eating an airframe every century or two.
Re:I'm not really understanding... (Score:4, Insightful)
Re:Government Contractors (Score:4, Insightful)
In the real world, whatever happens, everyone will pay for this. What do you think happens if that firm is properly insured? The insurance company pays and will increase rates for everyone, not just that firm that made the mistake (you can't do stats on a single mistake anyway, and the insurance firm needs to get that money from somewhere if they are to remain as profitable).
So everyone pays more insurance, this means the companies who pay more insurance have more costs and increase their rates etc. This is not something insulated. Ditto for bankruptcies, not everyone pays as much everyone pays for it in the end.
Re:Forgotten Lesson of WWII (Score:5, Insightful)
A often over looked factor is attrition in WWII. Made up numbers:
Lets say the US had zero elite level tankers but millions of noobs and we didn't start the land war until, well, frankly pretty much d-day 1944. Solution, make millions of noob-tanks. We didn't have any elite combat veteran tankers anyway to make use of elite level tanks.
Lets say the Germans had a hundred thousand elite combat vet tankers, but a quarter of them die in combat every year starting in 1939, so by 1945 you've got 12 year olds with hunting rifles "defending" Berlin at the last stand. Solution, make tens of thousands of elite-tanks and hope each elite-tank blows up more than 10 noob-tanks. Eventually you end up with dudes from the assembly line trying to be tankers, that didn't work out so well.
They darn near won, despite the attrition, so I wouldn't harsh their strategy too much.
Re:Even cheaper (Score:5, Insightful)
It would have prevented this disaster...
Re:Shit Happens (Score:5, Insightful)
Aircraft aren't cars. The moment you start treating them the same is the moment you sign your own death warrant.
Reading carefully in the article, the Air Force states that it is beyond economical repair, which usually means that the hours on the airframe are probably beyond some limit for stress or flight hours and to make such a huge repairs near the spar, which is the huge chunk of metal that keeps the wings on, would most likely require a huge program of testing, inspection, and re-certification.
Since the Air Force has dozens of spares of this particular airframe, it is more economical to pull a newer one out of storage and move all the stuff that makes a JSTAR a JSTAR to a new plane.
Re:Shit Happens (Score:4, Insightful)
It was not designed that way,
It was designed to bleed off excess pressure. The plane tried to do just that. Only some fuckwad left it plugged.
Let me come over to your house and install nails in your fuse fuse box and then tell you how badly designed your electrical system was that a simple short in an electronic device burned down your house.
It seems to me that since you put no thought into this that "blaming the bigger guy" is your thing.
Good luck with that.
Re:Shit Happens (Score:5, Insightful)
The airframe is a 40 year old ex-airline 707 with about ten zillion hours on it. A better analogy would be that it's like a $900 car with an $20k Oracle server in the trunk, and frame damage that would cost $2k to fix.
Re:Shit Happens (Score:5, Insightful)
No the professor was right.
There is a benefit to putting in the bolts the way the worker was taught to do it. It is also the standard way.
And the class came up with a number of solutions that would have been better than the upside down bolt.
You should always make assembly errors as unlikely as possible. Having a design that will fail if a single bolt is installed in that standard way vs a special procedure is just asking for trouble. Doing when other solutions are available is a fail.
Re:Shit Happens (Score:2, Insightful)
Assembly line work is mind-numbingly boring: install one bolt, over and over, for an entire run of the line, entire shift, or possibly for entire weeks or even months until they come out with a new version of the product. Assembly line workers don't have a half-dozen different things to keep track of... they have one thing, and they damn well better do it correctly.
Thinking is not one of their job duties. They are paid to do what they're told, not to think. If they put the bolt in the way they think it goes instead of they way they were told it goes, they are not a suitable assembly line worker - full stop.
Re:Shit Happens (Score:4, Insightful)
This is true *but*:
There is a hard rule in aircraft assembly that the bolt be placed head up nut down. This is to protect the plane if/when a nut falls off, gravity will still hold the bolt in place, hopefully long enough to land, or at least to eject.
This is not an optional rule, and assembly workers have it drilled into them at their new hire instruction, and every annual refresher, and whenever someone sees a mistake in QA, and just because someone thought now would be a good time to bring it up again.
It is "how it's done". <- full stop
Now, in this particular case, a dumbass engineer decided to have the bolt installed in contravention of this hard rule. He chose this because in the other orientation there was an issue with control cables, and for whatever reason the following options were not viable: move the bolt hole, use a shorter bolt, re-route the control cable.
The worker put the bolt in the way that she is "supposed to always" install bolts. Naturally this was not the right way for this bolt, and she is not blameless, but she is also not to be blamed for the entirety of the issue. She should have called her supervisor over and complained that the design conflicts with her training. Then put the bolt in upside down when her supervisor tells her to "just do it, will ya".
I still refuse to hold her as the sole cause of the issue. I've had people where I work refuse to do something against the "always do it this way" kind of rules and a design calls out something against that. 9/10 times we kick back the design as invalid. 1/10 we end up doing it, but only after everyone on the team has been trained that this one widget goes in wrong, and why. It prevents the "I know better" issues with people.
To sum up: Just because you don't pay them to think does not mean they will not think. Better to explain to them why things are done wrong in a particular case, then they will understand that it is not a stupid mistake that needs correcting, but rather a design tradeoff that had to be made.
-nB