AVweb

« Back to Full Story

The Dreamliner's Battery Fire

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

I'll bet the broadband video conference lines between Boeing, Yuasa and Thales are getting overtime use this week following Monday's ground fire in a Boeing 787-8 Dreamliner at Boston's Logan airport. The instant I saw the video of it, with smoke curling out of the after hatch, I thought to myself, that better not be a battery issue. But it now appears that this is exactly what it was.

This is serious stuff for Boeing and could put a significant dent in the airplane's certification basis if they can't figure out a solution. When Boeing proposed the Dreamliner a decade ago, it spec'd lithium-ion batteries for the airplane's innovative electrical power conversion system, which is made by Thales, with batteries from Yuasa. The airplane can also use conventional NiCad cells but either way, batteries provide starting energy for the APU and back-up power for other systems. At the time, the hazards of Li-ion were known, but Boeing wanted the weight savings and power density that only Li-ion can offer. Working with the FAA, it developed new certification standards for batteries under so-called special conditions. That required sophisticated electronics to provide overcharge and charge imbalance protection and physical thermal containment. Taken together and on paper, those precautions look more than robust enough to meet the stringent safety requirements of a modern airliner.

Well, maybe not. When I was researching Li-ion batteries for aircraft last year, I heard real concerns that the Li-ion main ship/starting battery risk wasn't yet well understood. As I reported, Cessna already lost one Citation to an Li-ion ground fire and withdrew Li-ion batteries as an option, at least temporarily. Now comes the 787 incident that might have gone the same way if it hadn't been caught early. It's also possible that the fire wouldn't have gotten any worse and that despite the smoke, the thermal containment worked. The airport fire crew showed up with the correct chemicals to fight an Li-ion blaze, but it still took 40 minutes to contain it completely. Early reports suggest the battery pack was well into progressive thermal runaway, with one failed cell torching off another. We'll see what the investigation reveals and I suspect it won't be long because Boeing isn't going to want the risk exposure if the problem is significant. The 787 is an ETOPS-330 airplane, meaning it can legally fly routes more than five hours from a diversion airport. That's a long time to have a smoke-filled cabin. Or worse.

The 787's flaws have been well publicized and the industry has made cooing noises that this is just normal teething pains that every airplane has experienced. That may be correct, although it's hard to compare the 787 to other airplanes because it's such leap forward in terms of all-electric sophistication. It's bound to have bugs. But I wouldn't consider an Li-ion battery fire a bug, so much as a confirmation that those who think Li-ion isn't ready for commercial airplanes could very well have a point.

Li-ion batteries runaway for three main reasons: They're sensitive to charging imbalance between cells, cells themselves short and external heat can ignite the highly flammable electrolytes. Once burning, they're hard to put out. Li-ion batteries are finding their way into vehicle markets, driven by the hybrid and electric vehicle push, but also as standard starting batteries in some new models. The record for these early adoptions seems good, both in performance and safety. But the Li-ion battery universe is still small. Flooded cells and AGM have been implicated in a few fires, too, but there are orders of magnitudes more of them out there.

I have five motorcycles in my garage, each with a blinking battery tender. I could replace the flooded cells with Li-ion, get more starting capacity for less weight, albeit at three times the cost. I haven't done that yet because I don't think the technology is quite ready for prime time and I don't care about the weight. Plus, I'm cheap.

Boeing obviously felt differently. I sure hope they're proven right.

Friday a.m. addition: The FAA is expected to announce a major review of 787-8 electrical systems and wiring complaint incidents.

Comments (41)

What about lithium-iron phosphate ? ( claimed to be safe)

If you are cheap, stick with ordinary lead acid batteries for all your vehicles. (Including electric cars) This point is made by Professor Richard Muller in his book: Energy for Future Presidents.

Posted by: Bill Berson | January 9, 2013 8:39 PM    Report this comment

The energy density of Li-ion batteries is a marvelous thing but at the same time their multiple paths to catastrophic failure make dealing with them a little scary, somewhat akin to carrying around a relatively well behaved but still powerful explosive.

One of the spookier aspects of these batteries is the way an internal fault in a single cell can sometimes fester for many hours or even days before cascading into full-fledged thermal runaway of the entire pack. There have been instances where EV or hybrid vehicles that were involved in accidents have had their battery packs abruptly go into runaway several days after the accident. I believe there are now special protocols for dealing with battery packs in electric vehicle accidents.

Posted by: John Wilson | January 9, 2013 11:22 PM    Report this comment

The batteries involved in the Citation fire were iron-nanophosphate technology. Yuasa is big in lithium cobalt-oxide, according to their web site. Higher overall capacity and watt per kg density. I saw one reference that said they may be using lithium maganese, but I can't find a confirmation.

Interestingly, Cessna encountered the first fire with about 50 airframes in the field and Boeing has delivered a similar number of 787s. Make no mistake, these batteries were regulated and certified to a fare-thee-well because of the risk sensitivity. I'm guessing there was an intense effort to think of everything and get this right.

Yet both have suffered fires with very little field exposure, hence the legitimate concern and fair question about whether the technology is mature enough.

Posted by: Paul Bertorelli | January 10, 2013 5:26 AM    Report this comment

Demonstrably, the sophisticated electronics to provide overcharge and charge imbalance protection and physical thermal containment, didn't. If you want an experimental determination you can conduct yourself, buy a lithium watch battery and, wearing eye protection, short it. It will shortly explode, with about the same result as a firecracker. First time I did this unintentionally, I was lucky I didn't lose my eyesight.

Posted by: Keith Peshak | January 10, 2013 5:28 AM    Report this comment

More than a few RC models and the odd workshop has been lost to Li-ion battery fires though with good precautions this seems to have gone down of late. Perhaps the batteries are better made as well.

Laptops and cell phones have been running on Li-ion batteries for many years now though there was a particular video on the web a few years back of a laptop catching fire during a lecture.

All these seem to have been put down to impurities in the battery construction - resolved with better build quality. Do we have a battery QC issue here?

Posted by: Graeme Smith | January 10, 2013 5:33 AM    Report this comment

I think we all need to get off our soap-boxes. As a pilot and avid model aviator I've been using LI-Ion batteries of all varieties since long before we had the knowledge and appropriate charging and discharging safeguards and procedures in place. In the early days, yes you did hear about a battery fire or two. Now they are basically unheard of, even with I expensive 'non-certified' balanced battery chargers and dischargers. It is very likely that this issue was caused by an electronics or programming error and not by a physical issue with the battery. Remember we are talking about an aircraft that is fly-by-wire. While I'd bet that the software that drives the charging/discharging of batteries is isolated from the software that drives the control surfaces, let's not forget the big picture - it's all software. My point: If you're going to worth about an issue, worry about the complete problem and address the root cause.

Posted by: James Keane | January 10, 2013 5:44 AM    Report this comment

Most of the comments, here and elsewhere, focus on the fire issue. Fire from lithium technology is easily dealt with. 'Thermal runaway', occurring INSIDE, can only be effectively dealt with by flooding with an water based liquid while isolating and preventing from spreading. My company, LithFire-X, has developed proprietary methods incorporating suppression, cooling, isolating and containment. The 787 story makes the headlines while most every pilot and passenger in a light plane is flying with multiple lithium powered devices!! My new IPhone five got very warm the day after I bought it. This issue will be with us as long as billions of these devices are used, stored and transported.

Posted by: Gerald Flood | January 10, 2013 6:26 AM    Report this comment

REALLY PAUL ? Really ? I happen to be a type rated 787 pilot. You'd think you might want to talk to someone who maybe flys one for living before you go spouting off like some kind of an expert. What will you pontificate on next ?

Posted by: Randolph Palma | January 10, 2013 10:12 AM    Report this comment

Well, Randolph, you're the expert. You tell us. I merely raised the question. Fairly, I think. We've now had two fires with Li-ion batteries, one a hull loss.

Posted by: Paul Bertorelli | January 10, 2013 10:18 AM    Report this comment

I always enjoy Paul's articles. He may or may not be an expert on some things, but I haven't read anything that makes me think he feels he's a 787 expert. I generally enjoy reading other peoples' comments about the topics he presents. It's too bad that some "experts" feel the need to clutter the comments with total lack of input except their need to tell the class how big their credentials are. Oh, my comment on the topic: 1) the 787 apparently has its share of development problems (ok, so do most complex projects), 2) lithium batteries have also had problems, and by now we know about their inherent risks, 3) I would hope the designers have more-than-adequate safeguards to prevent a real catastrophe 4) I just installed a lithium battery on my motorcycle.

Posted by: Eric Nelson | January 10, 2013 11:22 AM    Report this comment

The horrific onboard fire and the subsequent tragic crash of the UPS 744F flight in DXB was suspected to have been caused by a shipment of inactive container of Li-Ion batteries. Most flight crews have watched in shock and awe the video of a burning laptop and its multiple re-ignition onboard a training device during yearly emergency training syllabus. No one has yet discovered an easy way to pull off the road and jump off a burning aircraft over the Pacific Ocean, cruising at FL410! Although sophisticated battery technology and extreme electronics are not my area of expertise, I too was concerned when I learned of the use of such dangerous power supplies or ideas such as liquid cooling of hi voltage cables in the Dreamliner. Especially in a "330 ETOPS" aircraft with all composite structure! Most of us who grew old with Boeings and other jetliner manufacturers have experienced and understand "growing pains." But so far, 787 adolescent has been more of a nightmare than a pleasant dream! Perhaps I should ask those gentlemen over at the board room in Chicago; how is that outsourcing thing working out for you now?

Captain Ross "Rusty" Aimer (UAL Retired)

Posted by: Ross Aimer | January 10, 2013 12:14 PM    Report this comment

Erik Nelson + 1.

Posted by: Andres Darvasi | January 10, 2013 12:51 PM    Report this comment

Cool off your batteries Capt. Randolph, the fact you are type-rated on the 787 doesn’t automatically make you an expert in battery technology; except if you also hold a degree in Electrical Engineering and have recently studied this particular technology. If indeed you are an expert you could perhaps enlighten us with your insights, instead of belittling the host of this blog. Or maybe you are right, what does Paul know about Li-ion battery since he flies a piper Cub with no battery of any type? Seriously, Paul’s concerns about safety of lithium battery in aircraft are shared by many people on this blog. There is nothing dogmatic in Paul’s comments, as you are trying to insinuate. I’m not saying Paul is not opinionated, but are we not all opinionated? I’ve been reading Paul’s blogs for many years and I like how each week he brings a new controversial subject for discussion between Avweb readers.

Posted by: Andre Berthet | January 10, 2013 12:58 PM    Report this comment

It really would be nice if we had an accurate account of the history of problems with these batteries, what has been done about it, what the status of the newer derivatives of these things is, what can be done to mitigate the dangers, etc., etc............ So far we just have a lot of anecdotal stories, some so calledd news stories and a lot of opinions. Until someone comes up with a good, factual report then I am staying away from them.

Posted by: larry maynard | January 10, 2013 2:36 PM    Report this comment

I am a bit confused with the different chemistry specifications and each one's proclivity to combust.

We have LI ion batteries, Li-ion phosphate, Iron nanophosphate etc etc. Can we lump all of these together in terms of fire danger, or is there a difference between the species?

Sometimes distinction is the better part of valor.

Posted by: A R Goldman | January 10, 2013 2:41 PM    Report this comment

I know no more than your average Joe about Boeing's implementation of battery technology in their 787. The only perspective I would share is this: oftentimes when maing design decisions, I've eschewed very desirable options, where the liklihood of an undesirable occurance is statistically insignificant. Not because I feared that they would happen, but because in the event that they did, the consequences would be unacceptable.

As a point of reference, NASA's Shuttle vehicle design included over 700 (736, as I recall) "criticality-1" items, in which any one "failure will result in the nearly-certain loss of the vehicle and crew."

The point is not "how likely is it that will happen?" It's "what are the consequences if it does?"

As I read it, Paul's point was that this battery technology, as implemented, may not be ready for prime time. In my admitted ignorance, I'm inclined to agree.

Posted by: Thomas Yarsley | January 10, 2013 2:46 PM    Report this comment

Thomas and others, here's a link to the special conditions notice Boeing agreed to for the 787. It will give you a broad feel for what's involved.

www.snipurl.com/263gyki

As for the various Li-ion chemistries, here's a comparison of the principle types:

www.snipurl.com/263h06p

As for stability and safety, it all seems to be relative. A company called A123 made the iron-nanophosphate cells involved in the Citation fire. They claimed this technology is the most stable and declined to provide details on why it burned. A123 went bankrupt three weeks ago and appears to be going to Chinese interests.

People in the battery industry I have spoken to worry that an Li-ion fire is so energetic and difficult to control that it represents a high probability of aircraft loss. Obviously, Boeing thought whatever risk is involved--which I assume they assumed was 10 to the 9th--was worth it.

Posted by: Paul Bertorelli | January 10, 2013 3:16 PM    Report this comment

What I really don't get is this: I get e-mails from people in the RC field--and a comment above--saying stop getting your panties in a twist. We use these batteries all the time and they rarely burn anymore.

Yet these two Part 25-approved aviation batteries, which you'd think are made the highest possible standards with the best overcharge and balance threshold control obtainable, have burned twice. If the price of getting this technology right, including charge control software, is risking another hull loss, possibly with passengers, then I think our collective panties need some attention.

I honestly think Li-ion will get there as a safe aviation power source. I'm just not sure that this has been demonstrated yet. At this point, I wouldn't put one in my airplane.

Posted by: Paul Bertorelli | January 10, 2013 3:25 PM    Report this comment

Most prudent RC modelers charge the lithium batteries in a fire proof container or they stand nearby and watch it as it charges. I didn't read any mention of fire proof container in Paul's FAA document link. My understanding was that a lithium fire can't be extinguished by normal means. Did the FAA require fire proof container or not?

Posted by: Bill Berson | January 10, 2013 4:46 PM    Report this comment

Paul:

The Special Conditions are pretty tight. I especially note the use of the word "preclude." I think I know what that means.... Can you get ahold of the demonstration-of-compliance documentation? That would get into methodologies, validation testing, etc.

Posted by: Thomas Yarsley | January 10, 2013 5:42 PM    Report this comment

I think the really scary thing is the comment that this plane is ETOPS-330 certified. Really ?? I would hope the FAA would take that back until they figure out what is going on here.

Posted by: Michael Corder | January 10, 2013 5:42 PM    Report this comment

Somehow, this controversy over the wisdom of Boeing's design engineering of the 787 electrical system, brings back memories of how Boeing was able to convince the FAA that its single PCU control for the 737 rudder should be acceptable, precisely because the odds of a catastrophic failure of that unit was so remote as to be impossible.

Not too much different than the arguments of Boeing, McDonnell-Douglas and Lockheed, about how non-plug cargo doors would be just as safe as the plug type doors, incorporated in the passenger cabin.

Posted by: Robert Boser | January 10, 2013 6:04 PM    Report this comment

Thomas, not sure if I can get that documentation. But I'll see if it's public. My understanding was that the battery containment was supposed to include a physical thermal barrier. I don't know if Boeing talked the FAA out of that or not.

The general rule of thumb, say my battery sources, is that cell failure due to internal shorts will happen once in a million cells--not batteries, but cells. That doesn't account for charge imbalance failures.

The batteries involved are twice the output at half the weight. Almost magic. But if you add good thermal containment, you lose some of the weight advantage against what appears to me to be unclear risk mitigation.

Posted by: Paul Bertorelli | January 10, 2013 6:16 PM    Report this comment

As Paul B. pointed out, every battery technology can create excitement. Lead acid batteries can leak acid. They give off hydrogen gas when overcharged and build sulfate when undercharged. The sulfate can short a cell, causing an overheat condition.

NiCad have been known to go into thermal runaway, hence the advent of AGM to replace them. Paul reports there have been problems with AGM too.

Li-Ion and NiMH computer and wireless device batteries have failed rather spectacularly, but I have seen some of these devices suffer stunning abuses and come away still working, so I'm impressed we don't have more failures than we do.

Posted by: Thomas Connor | January 10, 2013 10:41 PM    Report this comment

The aviation industry is all about MONEY and weight savings. The money takes center stage every time weight savings is a distant second. Run the cost of lead acid cost /weight against Li-ion cost/weight.

As in all new technology we have to go through the growing pains and a few losses are part of the risk. History shows that older jets fell out of the sky, do they do the same today?

Well done Paul good subject

Posted by: Bruce Savage | January 11, 2013 5:11 AM    Report this comment

Below is a link to the FAA's listing of battery fires related to air transportation. This has been updated and some events removed, by the looks of it.

www.snipurl.com/263n5wo

Most of these are Li-ion batteries, but there are gel cells of various kinds in a few. Important to understand there are millions of cells out there so the incidence is still quite small. One wonders if it's a matter of time before an airplane is lost due to these fires.

All of the coin cells--CR123s and the like--fail in a characteristic way with a pop or fiery explosion. This, I'm told, is because they contain metallic lithium, which is quite volatile.

Posted by: Paul Bertorelli | January 11, 2013 6:33 AM    Report this comment

In the world of RC modelling no-one uses Lithium-Ion batteries anymore as they are so problematical eg burnt out planes, cars, trailers, garages, houses etc. Li-Ion bats have been largely replaced by LiPo, LiFe as they are much safer. Why does the aviation industry insist on using old intrinsically unsafe technology?

Posted by: Mike Smith | January 13, 2013 1:30 AM    Report this comment

Mike, both lithium polymer and lithium iron-nanophosphate are lithium-ion technologies. They still have lithium salt electrolytes made of flammable organic solvents.

And they're still sensitive to charging imbalances. even though they are considered to be more stable chemistries. As I noted, the Citation that burned had LiFEPO4 technology.

Posted by: Paul Bertorelli | January 13, 2013 5:59 AM    Report this comment

I've got a fair bit of faith in Boeing and the FAA but maybe these cells should be jettisonable in flight? A bit like the panel-mounted red button I'm working on that ejects annoying passengers. I can email Boeing my rough sketches to get them started.

Posted by: john hogan | January 13, 2013 6:52 AM    Report this comment

Hello Randolph Palma, your post accusing Paul of not knowing anything because he might not have consulted with you are another type rated 787 pilot was disgraceful. As a fellow commercial rated pilot you reflect poorly on all pilots. You should know the serious nature of on board fire and that this batteries indeed are catching on fire and a hazzard. Furthermore, you demonstrate poor CLR by being publicly rude and arrogant by belittling others. Do you also belittle your flight crew members? If you make them feel like crap too, you are more a flight hazzard than any battery. It takes a team to operate safely. I would love to hear about your background and your in-flight behavior seeing you publish such a disgraceful and berating comment in public! Did you also "scab" at continental Airlines since they are the only U.S. operators of this aircraft and full of scabs in the senority range that would allow flying of this machine?

Posted by: Bob Lee | January 14, 2013 11:33 AM    Report this comment

I saw a photo of the burned battery box last night on King 5 news ( Seattle). It looked badly scorched, but mostly contained, with burn marks where flames exited through various ports in the box.

Posted by: Bill Berson | January 15, 2013 7:44 PM    Report this comment

Another report of a possible 787 battery fire, again. I'm a pretty big Boeing supporter, but even I'm getting concerned at this point. I think on the whole, the 787 can live up to its reputation to revolutionize air travel, but only if they can get these battery/electrical issues under control, and soon.

Posted by: Gary Baluha | January 16, 2013 2:40 PM    Report this comment

The FAA has issued an emergency AD today grounding the 787 until the battery/s is made "safe".

Posted by: matthew wagner | January 16, 2013 6:52 PM    Report this comment

"I saw a photo of the burned battery box last night on King 5 news ( Seattle). It looked badly scorched, but mostly contained, with burn marks where flames exited through various ports in the box."

This is actually good news, because it means the thermal containment may have worked as it was supposed to. But the very bad news is with only 50 airplanes out there, the 787 appears to have had two battery fires. This is wildly unacceptable risk for a modern Part 25 airliner, as I'm sure Boeing would be the first to admit it.

Going in, both they and the FAA knew of the Li-ion exceptional risk, which was the reason for the special conditions on the cert. Yet with all that scrutiny, the best this effort can manage is two battery fires in a fleet of 50 airplanes? The FAA was right to ground them. I wouldn't be surprised to see an emergency AD converting them to NiCads until the Li-ion technology smooths out.

Posted by: Paul Bertorelli | January 17, 2013 5:00 AM    Report this comment

In a post above, I said: "The point is not 'how likely is it that will happen?’ It's 'what are the consequences if it does?' "

It appears increasingly likely that Boeing is suffering from consequences of a design-decision situation like the one that I was describing.

I've had an inextinguishable in-flight fire. The only non-fatal solution was an expeditious landing. In an ETOPS 330 operation, that option often is unavailable. Classic “unacceptable consequences.”

Let’s hope that we're not about to witness a case of the tail wagging the dog: superhuman efforts to make LI battery technology do something that it perhaps simply cannot do – simply because the weight and volume characteristics of alternative battery technologies may be incompatible with the existing airframe configuration.

Not to sound alarmist, but this situation reminds me of the day after the Apollo 204 fire. The consequent re-design cost the program almost two years’ time – but it was the right thing to do.

My Spidey sense is tingling...

Posted by: Thomas Yarsley | January 17, 2013 7:22 AM    Report this comment

On King 5 news ,The retired Boeing Li-ion battery engineer said: (paraphrased) "it did not puff up or explode,so no big deal" and "lithium is green"

It might not bring down the plane, but smoke in the cockpit is still a big deal, I think.

Posted by: Bill Berson | January 17, 2013 10:12 AM    Report this comment

I am soooo glad that I'm not working on the 787 anymore. Granted I wasn't involved in the batteries, but I can only imagine the hell the group I used to be in is probably experiencing.

Posted by: Andrew Upson | January 17, 2013 3:42 PM    Report this comment

'Lithium is green" Well there ya have it folks, hippie bureauspeak that translates a self-igniting rare-earth into a warm fuzzy. I used to work on the Boeing E3 AWACS program as a test and evaluation director and could just imagine the spitwads and harangue we'd direct at the project manager if he made such a statement. How times have changed!

From my experience with the E3, Boeing does very little that is stupid, preferring to test stuff to death before release to the customer. But the customer was rarely so patient, often asking to assume risk to expedite fielding an item or subsystem. One wonders if that happened with the battery issue?

It would also be useful to know what the weight savings of li-On is compared to alternative batteries.

I also assume that since the battery(s?) are a high-attention item they are instrumented every which way and the engineers have data that show temp, pressure and escape velocity compared to some well abused batteries in the lab.

Posted by: Thomas Connor | January 17, 2013 5:13 PM    Report this comment

part 2: Then there is terminology. 'Fire' evokes a lot more emotion than 'smoke,' and fumes usually less. Odorless, colorless gasses like carbon monoxide get a yawn from Joe Sixpack. Never mind that what can kill is usually in the reverse order listed.

One wonders where the word 'fire' got injected into this discussion. Has anyone reported seeing flame or are we mis-characterizing an overheat condition the same as flames licking up the stairwells? I say that because smoke, fumes and gasses can be vented but a fire needs to be fought.

Normally, an overheat condition is also easy to resolve. Having a battery behave like a piece of pissed-off plutonium is a new one for most folks, but I understand NiCad thermal runaway is not uncommon in jets, so one wonders if this is the same movie with a different title?

Posted by: Thomas Connor | January 17, 2013 5:14 PM    Report this comment

The next issue here is the composite (thermoplastic) inherent problems. The B787 has more than 60% of its mass making it more critical than on the Airbus A350 or A380 plastic structure. The same disdains was given by Boeing and Airbus when the plastic burning, breaking and aging effects were questioned by several in-house design engineers. Some were fired for bringing this to the table. The FAA, well they just follow orders. Plastic ages becomes brittle and breaks, it is more complicated to reinforce the composite structure than to replace battery systems.

Posted by: Rafael Sierra | January 24, 2013 2:33 PM    Report this comment

Does anyone have any more information about the Cessena battery incident? I don't find anything anywhere.

Posted by: Gerhard Randers-Pehrson | January 25, 2013 11:48 AM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?

Register

Enter your information below to begin your FREE registration

« Back to Full Story