Boeing 787 battery fire

[ls99]

Food for thought:


Now, investigators at the National Transportation Safety Board have determined the cause of the battery fire in the Japan Airlines 787, saying that the fire was caused by a short circuit in one of the battery’s cells leading to a thermal runaway which led to the other cells catching fire. The NTSB said that temperatures inside the battery exceeded 260 degrees Celsius (500 degrees Fahrenheit).


Ars Technica
Investigation reveals cause of battery fire on Boeing 787 Dreamliner

Edit: Headline should spell Boeing.

 

[Texas Proud]

I kind of did not believe those first reports when they said the batteries were good... I think they had not tested them the way they were being used..


But hey, what do I know about it.... it was just an opinion....

 

[growing_older]

There were reports of similar battery fires on test benches in 2006, including one that burned down a facility A 2006 battery fire destroyed Boeing 787 supplier. The whistleblower lawsuit on that was dismissed not on merit, but on procedure.

Boeing was also able to set their own safety standards for these batteries because their use in aviation was new, then grandfathers them in after government standards were written in 2008. This sounds like a regulatory mistake. Boeing were allowed to carry out their own safety checks on Dreamliner jets as there were no federal guidelines for testing lithium batteries that grounded the fleet | Mail Online

I wonder what corporate pressures lead to including such a risky component in a new airplane design. It seems very unlike past practices.

 

[samclem]

This is very good news for Boeing. The investigation into these fires was seemingly stuck in a dead-end with no good leads, a situation that could have lasted for months. It's tremendously expensive (in money and reputation) to have these new planes sitting around, and if investigators have truly identified a likely cause then it clears the way for a solution.
It may seem obvious, but there's hardly anything scarier in an airplane than a fire. There's no getting away, and there may be very little time to resolve the issue.
I'm sure Boeing's fix will include more than just internal changes to the batteries or better QC at the battery mfgr. I've read of some type of containment around the battery to reduce risk if flame spread and also of improved detection systems.
Anyway, I wish them a speedy, thorough fix to the problem and a return to flight for the 787s.

 

[ls99]

An explosion and fireproof locker would be an improvement of the battery's current location:

 

[samclem]

787 Fire at Heathrow

No word yet from Boeing or the airport on the cause/location of a fire today aboard a 787 at Heathrow. It might have nothing to do with the batteries (could be a fire in the galley, etc.)

Story

Boeing shares were immediately down 6% on the news.

 

[FIREd]

From the pictures of the damage on the BBC's website, it looks like the fire was located in the back of the cabin, just in front of the tail fin. Pretty far from any battery compartment AFAIK.

 

[BigNick]

From the pictures of the damage on the BBC's website, it looks like the fire was located in the back of the cabin, just in front of the tail fin. Pretty far from any battery compartment AFAIK.

I thought the battery was in the tail on the 787?

 

[FIREd]

I thought the battery was in the tail on the 787?

Based on what I read online, the main battery is located in the front of the aircraft (~under the cockpit), while the APU battery is located under the main cabin, somewhat aft of the wing box.

 

[Dimsumkid]

I thought the battery was in the tail on the 787?

From the article:
But the main electrical panels and generators are in the center of the plane, below the passenger floor. That is also where one of the two lithium-ion batteries is situated. The second is near the front of the plane, under the cockpit.

 

[growing_older]

Whether the latest incidents with 787s are related to batteries or some other problem, this does seem to be a plane that [-]broke a lot of basic design practices[/-] pioneered new outsource methods, and it's unlikely these are the last problems that will surface. I am not convinced that the battery problems are fixed. I am not convinced that these new problems are the only other ones that will be found. Good engineering seems to be in conflict with some corporate needs and engineering doesn't seem to be getting as much support as they should. There's a awful lot of money at stake in these delivery schedules.

 

[Ed_The_Gypsy]

Good engineering seems to be in conflict with some corporate needs and engineering doesn't seem to be getting as much support as they should.

Boy, did you hit the nail on the head! True in my industry(s), for sure. There is a lot of wishful thinking at the executive level that can result in serious losses of all kinds. Think Challenger.

 

[ERD50]

Boy, did you hit the nail on the head! True in my industry(s), for sure. There is a lot of wishful thinking at the executive level that can result in serious losses of all kinds. Think Challenger.

I just read one of Feynman's books ( Amazon.com: "What Do You Care What Other People Think?": Further Adventures of a Curious Character (9780393320923): Richard P. Feynman, Ralph Leighton: Books ), and it was pretty amazing (and scary) how management talked themselves into supporting unsupportable numbers (see below).

I saw some of that at MegaCorp, but lives were not at stake (livelihoods, yes).

I was also wondering of the parallels with this Boeing Dreamliner. I assume they have lots of very competent engineers, I assume there are lots of tests and simulations being done. I know this is a complex beast, with lots of new applications of newer tech, but this seems like a lot of problems this late in the game. I do wonder if engineering was pushed too far?

OTOH, if you don't push engineering, you would never ship anything - they (we) always want another crack at improving it and checking it and improving it again. Gotta strike a balance.


From Fenyman: Management said the chance of a shuttle complete failure was ~ 1 in 100,000. Fenyman pointed out that meant that you could launch a Shuttle everyday for 274 years and have only one complete failure on average. That does not pass common sense. When he interviewed the engineering teams on several of the major sub-systems, they estimated something like a 1 in 200 chance of complete failure of their subsystem alone. Multiply that by the various subsystems, and you lower the reliability figure. Management's number was a 'wish', not a calculation, and innocent people rode the shuttle based on that 1 in 10,000 number.

-ERD50

 

[check6]

If it had been made of aluminum, not sure the hull would have been penetrated as shown in the photos. It is going to be really interesting how they repair a carbon structure. My company was a lead customer for this plane and smartly canceled their order when delays first began. Personally, I have no interest in riding in one.

 

[martyp]

Good engineering seems to be in conflict with some corporate needs and engineering doesn't seem to be getting as much support as they should.

"Good engineering" necessarily involves compromises regardless of profit motives. Everything can't be tested. Time does matter and, yes, costs matter. If a product or system can't be delivered at a reasonable cost and at a reasonable time when the customer needs it (or can use it) then all the engineering in the world is pointless.

Clearly lives are at stake with a product like an airplane but the standard is "best engineering practice" when a new complex technology is being used. When you are doing something for the first time there will be no existing standards. You extrapolate from what standards are out there (if any) and then you are making it up as you go along. Along the way you do risk analysis but there will alway be risks.

Having provided engineering support for a production line I can say that there are just some things you can't test for until you put it into full production. Particularly low probability risks.

There are lots of technologies that we use that have life and death consequences for failure. I'm sure most have pushed the limits of past experience. There is such a thing as gross negligence and if that is the case here then the company should suffer. And it is not just management that can make bad choices. Engineers can become overconfident as well.

 

[growing_older]

Boeing had well developed "good engineering" practices which dealt with all these conflicts you describe. For the 787 they abandoned many of those in pursuit of new business partnerships which meant some core engineering was done by other firms outside of their usual controls using other practices and there was tremendous time pressure to accept whatever was delivered. Even if the panels didn't fit together. Even if the battery supplier's facility was burned down by unexplained battery fires. Even if Boeing engineers objected to work that was delivered.

Yes, there is a check and balance between "good" engineering and reasonable corporate objectives. The 787 program doesn't seem to be an example of that.

 

[ratto]

Per this article, Boeing: Cause of Dreamliner Fires 'Almost Doesn't Matter' | National Legal and Policy Center, the top engineer at Boeing, Michael Sinnett, told reports at Tokyo a few months ago:

the enclosure keeps the airplane safe, it eliminates the possibility of fire, it keeps heat out of the airplane, it keeps smoke out of the airplane, and it ensures that no matter what happens to the battery, regardless of root cause, the airplane is safe

In some ways it almost doesn’t matter what the root cause was.

What could be possibly better than this?