Slashdot Mirror
Japanese Probe Finds Miswiring of Boeing 787 Battery
NeverVotedBush writes in with the latest installment of the Dreamliner: Boeing 787 saga. "A probe into the overheating of a lithium ion battery in an All Nippon Airways Boeing 787 that made an emergency landing found it was improperly wired, Japan's Transport Ministry said Wednesday. The Transport Safety Board said in a report that the battery for the aircraft's auxiliary power unit was incorrectly connected to the main battery that overheated, although a protective valve would have prevented power from the auxiliary unit from causing damage. Flickering of the plane's tail and wing lights after it landed and the fact the main battery was switched off led the investigators to conclude there was an abnormal current traveling from the auxiliary power unit due to miswiring."
Who will it be? Maintenance? Boeing?
So basically, the user reached back behind the power supply while fiddling and bumped the 110/220V switch, and it caught fire. Naturally, they didn't say anything to the tech after setting the switch back besides, "It just caught fire! All by itself!"
The user in this case is a giant airline company, and tech support would be Boeing. The FAA, of course, is the QA manager, who reviewed the call, and after reading the ticket closure notes, facepalmed, leaned back into his chair, and took a deep draft of coffee.
#fuckbeta #iamslashdot #dicemustdie
Can't they make an idiot proof power plug?
Because idiots are much more resourceful than ordinary people.
#fuckbeta #iamslashdot #dicemustdie
Yes, but commercial airliners aren't built with plugs and sockets. For weight savings, everything is directly hardwired. At least, in pretty much every airliner prior to the 787, and I can't imagine Boeing changing that. Military aircraft are built with plug and socket connectors, but both sides of the connection are big bulky heavy metal components. When your plane is a cockpit and wings strapped onto a giant oversized turbine, you basically don't care about weight, but commercial airliners are the exact opposite. They're obsessed with weight savings, so the miswiring happened during initial assembly and their quality control procedures were too poor to catch it. Boeing has fallen a loong long way.
Someday, people are going to look back on the outsourcing mania of core competencies by MBAs over the past generation as sheerest idiocy.
Because idiots are much more resourceful than ordinary people.
Behold!
It's the damned metric +/- that causes all the confusion.
"the battery for the aircraft's auxiliary power unit was incorrectly connected to the main battery that overheated, although a protective valve would have prevented power from the auxiliary unit from causing damage"
What is a power diode
AccountKiller
The Japanese are doing everything they can to point blame away from Yuasa. As someone above noted, and, yes I did read the fine article, they have concluded that it must have been wired incorrectly after seeing some lights flicker and a white dove flying in the southern quadrant.
Yes, but commercial airliners aren't built with plugs and sockets. For weight savings, everything is directly hardwired. At least, in pretty much every airliner prior to the 787, and I can't imagine Boeing changing that. Military aircraft are built with plug and socket connectors, but both sides of the connection are big bulky heavy metal components
Do you have a reference for that? It doesn't make sense that field replaceable parts are hardwired in - you'd have to clip the wires to take it out, and every time you clip the wire it gets shorter, so eventually you'd have to run a new wire back to the source.
Even for parts that aren't replaced often, it seems that hardwiring would just increase the chance of error - if everytime they replace an engine someone has to sit down and manually splice 200 separate wires, that seems a lot more trouble prone than plugging in a dozen connectors that were wired in at the factory and tested on the factory test harness to be sure every wire was connected to where it should be.
What the fuck are you talking about? There are cannon plugs all over the place. The reason you think "directly hardwired" is because almost each LRU has its own circuit breaker. You must be one of the fucking PHBs
It's gonna be fun to watch. Whistleblowers are starting to emerge telling of how safety issues were swept under the rug in the name of getting an already late project out, and engineering concerns were ignored or overridden by managers. It's very similar to the story of the O-rings on the Space Shuttle, where engineers knew there were problems, but their concerns were ignored.
Gonna be a real show over the next year as these people start coming out in greater numbers.
When you say "Japanese Probe" I had an entirely different idea in my head regarding what this story was about.
Slightly different, but my friend works for a company that makes in-flight video systems for planes including Lufthansa. While not mission critical, they still have to follow FAA and other regulations... one of which is some of the plugs they use plug in and then are secured in place with 12 to 16 screws even though the signals being passed are just network/video/audio.
I don't see why they couldn't use plugs of the same fashion instead of hard wiring everything
What in the world are you talking about? Very few connections are hardwired, almost everything has a plug, usually, a rotary cam-lock plug from Cannon, etc.
Do you think they solder all the avionics in place?
Those "idiot MBAs" landed a comfortable retirement plan.
The "perfectionist engineers" were laid off.
What used to be the crown jewels of the company, the core competencies, are now commoditized for lowest cost overseas manufacture.
Most of our core corporations have been reduced to a cadre of highly paid salesmen selling a branded product, of which one can purchase identical generic equivalents. The last vestige of holding onto power is by keeping the generics out of the market by legal maneuverings of copyright and patent law.
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
I'd say in-flight video is mission critical, look at SwissAir 111. The company was so eager to cash in on midair gambling they overloaded the wiring with standalone power supplies rather than putting in a more sensible system. Crash. http://en.wikipedia.org/wiki/Swissair_Flight_111#TSB_findings
For the Prius accelerator screwup....
AFAIK you are completely incorrect. Just using the 787 for example, the entire thing is built in modules including wiring and all, that are built and tested by subcontractors and then plugged and bolted together at the Boeing assembly plant. All commercial aircraft that I am aware of at least since the 1940s has had connectors. The biggest problem with connectors is not the weight but the unreliability. Each connector is a potential point of failure, so aircraft electrical connectors are actually heavier - they have positive scraping between the two parts of each connection, often have moisture resisting / sealing. and have a threaded ring that holds them together. Then (IIRC) there is an additional set of tabs through which a wire is threaded and itself positively bound - used to be twisted, now I think they use a mechanical crimp. The wire assures that the threaded ring can not unscrew itself due to vibrations.
The most expensive cost for a commercial aircraft after fuel is the cost of downtime - time spent fixing things costs thousands of dollars per hour. Therefore everything on an aircraft is designed to be removed and disconnected quickly, efficiently and safely - including things like wings, tail fins, etc. FAA is not going to allow the mechanics to cut wires and fasten them back together, so again the connectors are designed so that each one can only go one way.
It's easier to be a result of the past, but more fun to be a cause of the future! http://www.spacefinancegroup.com/
I'm sorry; but, you're wrong. Work for a sensor manufacturer that sells to the aerospace industry and I can tell you, commercial aircraft cabling is full of connectors. Same kind of locking connectors found on military aircraft.
Yes, but commercial airliners aren't built with plugs and sockets. For weight savings, everything is directly hardwired.
So they've managed to skim off maybe 10 pounds off the design of the aircraft, saving some several thousands in fuel costs over the operating life of the aircraft. A reasonable tradeoff considering the chance of the aircraft catching fire and then exploding when it hits the ground, killing everyone on board. *sips tea* Yeah. Makes sense to me. I mean, what's the cost of settling an accidental death claim for 300 people?
Here comes the math!
The cost of failure:
A product defect typically weighs in at $2.1 million USD per. So assuming 300 passengers and 10 crew, that's $651 million payout per plane going pop.
The cost savings:
Now, a 747 at least uses a gallon of fuel per second, or about 5 gallons per mile (average) on a flight. A typical domestic flight is about 2.5 hours in flight time, or 9,000 gallons of fuel. The weight of the aircraft, empty and unloaded, is about 95,000 pounds. It has 171 miles of wiring. Let's assume that we want to add connectors every 100 feet; That gives us 902,880 connectors. The average weight of a connector we'll say is 1.5 grams. that gives us 1,354,320 grams of extra weight to add connectors, or about 25,031 pounds.
So to add all those extra connectors would add an extra 26.3% cost to fuel. Now, the Dreamliner is slated to have a service life of about 30 years. We don't know how many pressurization cycles that equates to, but we can make an estimated guess. Let's just say 2 flights per day, 5 days a week. That'll be 1,560 flights before retirement then.
The average domestic flight is around 700 miles, we'll say. If the fuel cost before modification is 5 gallons per mile, at $3.30 per gallon... the cost of fuel per flight is $46,200. With the modification, it would cost $57,750.
Fuel cost over life of vehicle (before mod): 72,072,000.
Fuel cost over life of vehicle (after mod): 90,090,000.
Difference: $18,018,000.
Cost on failure: $651 million
Failure rate cutoff: 1 in 36
In other words, if a catastrophic failure that could have been prevented with electrical connectors happens more than 1 out of 36 planes, it's worth it. Otherwise, it's not.
#fuckbeta #iamslashdot #dicemustdie
The original
Questions raise, answers kill. Raise questions to stay alive.
Hmmm -- a quick bing search for "aviation grade connector" shows lots and lots of connectors. There are even magazine articles about them.
http://www.aviationtoday.com/av/issue/feature/Product-Focus-Connectors_18865.html
Want a sig like mine? Join ACM's SigSig today!
Posting to undo an egregiously stupid mod. #spamapology
Do not mock my vision of impractical footwear
Presumably the same person who miswired the Japanese 787 miswired at least one other plane. That should be easy to check. If there are other planes with smoking batteries, check the wiring, then do it for all other 787s.
In a time of universal deceit, telling the truth is a revolutionary act. George Orwell
Nikkei is reporting that this miswiring, which connected the auxiliary and main batteries by mistake, is due to a bug in an older version of the 787's electrical design. The bug was found and fixed in November 2011, but somehow Boeing never got around to re-wiring this particular aircraft according to the new plan. Nikkei also reports that the miswiring probably did not cause the battery fire, because there is some sort of an "anti-reverse current mechanism" in place to prevent damage even if the auxiliary had started to feed current into the main battery.
Someday, people are going to look back on the outsourcing mania of core competencies by MBAs over the past generation as sheerest idiocy.
I think my grandad said the same thing about transistors. You'll rue the day you ever gave up on valves he'd yell...
Re: The battery cells are rectangular with a stud on each side of the top. Not even any prominent markings to indicate polarity, though the two studs seem to be mounted with different colored rivets.
.
The other possibility is that the installer was color blind and has been able to get by without that disability showing through. Most items that are color-marked often have a redundant marking that is not dependent on color vision perception (except for resistors and their color banding indicators, http://en.wikipedia.org/wiki/Electronic_color_code#Resistor_color-coding and for cable runs of twisted-pairs that use paired coloring indicators
You need a blow-out panel.
The M1A2 Abrams tank has one for the shells. If they start to burn, the blow-out panel pops off and the whole mess exits the tank.
Factories that make vinyl have them. When the concoction goes boom, blow-out panels prevent total destruction of the building. Workers may even survive.
Meth labs don't have them. :-)
A reasonable design would have several battery compartments, each with a separate blow-out panel. These should be located so that debris will not enter the engines or get run over by the landing gear. The rear underside seems like a good location.
This reminds me of when I was having a new furnace installed. The installer wiring up the thermostat was having a hell of a time because he was color blind (I helped him out there).
Looking at the pictures here though even a color blind person should be able to easily tell positive from negative on the battery terminals.
This post is encrypted twice with ROT-13. Documenting or attempting to crack this encryption is illegal.
Plugs can also have issues when compared with color-coded cables. An Airbus A320 had a scary situation when the cable to the captain's side stick was inserted incorrectly into the plug by the mechanic who maintained it. Left and right were reversed and the crew missed it when performing the checklist and thus only noticed a problem when the aircraft had just become airborne. Luckily they were able to quickly identify that there was a problem with the captain's side-stick and the first officer pressed his override button and thus they could climb without the passengers ever realizing what a near-miss it was and were surprised when they had to return immediately due to a technical problem. It did lead to the plug being redesigned to be impossible to insert incorrectly but I doubt that that lesson has propagated everywhere in the industry (not that I'm pointing at Boeing in particular here since I think all manufacturers still probably have similar flaws in existing aircraft, which is why mechanics are worthy of more respect than they usually get because nobody appreciate mechanics when they do their job right and nothing happens).
Never underestimated the ingenuity of dumb. I gave a guy a hard drive to replace the one in his computer. He 'knew all about it' and I knew he had the knowledge to format and reinstall his OS no problem.
He brought the drive back to me, claiming it didn't work, in fact the power cables would not fit till he took his dremel to them, so it must not have been the right drive for his machine.
He had shaved off the corners of a standard hard drive power connector so he could fit it upside down in a used 512 MB IDE hard drive. Which of course killed the drive, but I also determined it killed that connector on the motherboard's IDE bus.
The reason for this mistake?
His computer had been built with some oddball brand of hard drive (I can't remember... Paladin or Palladium? I think it started with a P) which put the controller board on TOP of the drive instead of on the bottom.
Um, you're off by two orders of magnitude. 171 miles / 100 feet = 9,029 connectors, not 902,880. So the failure rate cutoff (assuming the rest of your calculations are correct) works out to 1 in 3600. Care to re-analyze?
Weeks of coding saves hours of planning.
The UK tried.
They ended up with the safest plug in the world... that causes more foot injuries than any other plug.
Behold the BS1363, bane of the foot.
Calling someone a "hater" only means you can not rationally rebut their argument.
And also: 1,354,320 grams is 2,986 pounds, not 25,031 pounds. (Correcting for this, as well as the number of connectors, makes the actual failure rate cutoff 1 in 30,180.) You didn't work on the Mars Climate Orbiter, did you? ;-)
Weeks of coding saves hours of planning.
Asymmetrical plug/socket design. It's not that difficult.
They sentenced me to twenty years of boredom
That kind of math only works if electrical connectors are the sole reason for possible catastrophic failures that can be blamed on the manufacturer.
Since there are also possible hydraulic failures, wiring fires and so on, the manufacturer certainly needs a buffer larger than one in 36.
"It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
I love those plugs. I've lived in many countries and I've frequently had trouble with sockets that don't reliably make contact, wall warts that fall out of sockets and sockets that spark like July 4. The British plug doesn't have these problems. It just works. And you soon learn not to leave them lying about.
No argument here. the 2 pin US and Euro plugs seem to fall out if so much as a slight breeze hits them. A shame these plugs are so prevalent in Asia.
The Australian plug is the same, once it's in you know it'll take force to remove it, but a bit less deadly to tread on.
Calling someone a "hater" only means you can not rationally rebut their argument.
Did they try turning it off and on again?
I see that the discussion here is based on a sketchy summary from the originally Japanese press conference. More coherent information is available if you could read Japanese but I know it's too much to ask for...
Here is the latest update of the on-going investigation from the JTSB issued 20 Feb, 2013; this mentions the mis-wiring:
http://www.mlit.go.jp/jtsb/flash/JA804A_130116-130220.pdf
More in-depth information is given at
http://www.aviationwire.jp/archives/16032
According to this article, the mis-wiring was in the original specs/design, and the design had been corrected. The aircraft in question was manufactured in accordance to the earlier specs but no modification was made to comply with the new ones. One can infer that the bug was considered insignificant to compromise the safety of the aircraft. The JTSB currently does not think this mis-wiring was the cause of the battery incident although they will keep looking into it as a potential cause of anomalous voltage readings.
Aircraft certification standards for material flammability were inadequate in that they allowed the use of materials that could be ignited and sustain or propagate fire. Consequently, flammable material propagated a fire that started above the ceiling on the right side of the cockpit near the cockpit rear wall. The fire spread and intensified rapidly to the extent that it degraded aircraft systems and the cockpit environment, and ultimately led to the loss of control of the aircraft.
I'd give my right arm to be ambidextrous.
The capaitors are connected in series regardless of how the voltage selector is set.
The ends of that series combination are connected to the AC supply by a diode bridge, they are also connected to the load. The switch is connected between the midpoint of the capacitors and one side of the incoming supply.
With the switch off (240V mode) the system acts as a normal bridge rectifier and the capacitors are charged in series during both half cycles.
With the switch on (120V mode) the system acts as a voltage doubling rectifier and charges one capacitor during each half cycle.
Either way you get about 300V DC across the capacitors which can be used to power a flyback converter to provide isolation and voltage conversion.
This design is cheap and works ok but connecting a rectifier-capacitor circuit directly to the mains results in a HORRIBLE power factor (even worse than connecting one via a transformer as the transformer limits the harmonic currents). As PCs have become both more common and more power hungry this horrible power factor has been deemed unacceptable (at least in europe) and all but the shittiest power supplies have moved to other designs
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register