I suppose this (assuming it was the batteries again. I understand they swapped them out with a new design) is the downside to making a system fail safe instead of making it just not fail. They modified the battery compartments so that a fire wouldn't threaten the air-worthiness of the aircraft but any fire is still likely to cause bad press and, in this case, disrupt traffic at an airport.
> Unfortunately, the pack architecture supplied to Boeing is inherently unsafe. Large cells without enough space between them to isolate against the cell-to-cell thermal domino effect means it is simply a matter of time before there are more incidents of this nature.
The question in my mind is far more fundamental than anything related to the design of the battery packs or whether or not Elon Musk's involvement is of any real engineering value.
None of that matters. He/they would have to convince me that LiPo/LiIon packs are safe. This coming from someone who's waiting for Tesla's SUV's. I kind of draw a demarcation line between flight and road-based vehicles. Fires are a thousand times more dangerous at 30,000 feet.
The fundamental issue I keep coming back to is that Lithium Polymer batteries are dangerous. One or multiple cells. It does not matter.
I've been using and abusing them extensively for, I don't know, maybe a decade, in all of my RC gliders and helicopters. I keep them locked-up in a fireproof container and never --ever-- charge them unless I can be in the room during charging.
There are plenty of accounts in RC circles of battery packs spontaneously combusting, catching fire in flight, catching fire while charging or lighting-up if damaged. The problem, as it was explained to me, is that these cells can produce methane and have an avalanche failure mode. Methane is a product of a manufacturing defect whereby a small amount of moisture is sealed into the package. Lithium reacts unfavorably with moisture and the rest is history.
Lithium Polymer battery fires are extremely hot, fast and very powerful. I've seen it myself with three and six cell packs. I would not want to be anywhere near a fire in a larger pack.
And so the fundamental question for me is simple: Why? Why use Lithium-based chemistry for a flight battery pack. Yes, I know, lower weight, lower volume. However, I am not sure these packs can ever be considered to be safe. How can an aircraft manufacturer ensure that the quality of the packs it assembles is as required? Can you ever know?
I know millions of Lithium-based batteries are in use in everything from phones to laptops and more. Not sure how to think about that. A fire is a fire. A laptop stowed away in cargo can still cause a huge problem if it catches fire.
To be fair, I have never had a spontaneous fire in any of my thirty of forty packs (ranging from 2 cells to 12 cells). I have had cells spontaneously puff-up --inflating like a baloon-- and destroy the pack by expanding so much that the battery pack's casing cracks open. In those cases I've used them to experiment by overcharging to see how they light up. It does take a lot, but when they do it is massive fireball.
Perhaps someone with information closer to the source or quantifiable safety and reliability data can pitch in?
EDIT: Prior to LiPo's I was using NiCd (Nickel Cadmium) cells and have never had any issues, even with abuse and mechanical damage.
"And so the fundamental question for me is simple: Why? Why use Lithium-based chemistry for a flight battery pack. Yes, I know, lower weight, lower volume."
And more power density. I'm hoping an aeronautical engineer will eventually chime in, but this document is a pretty good technical read:
The major change was going to a "no-bleed" engine setup which allows for more efficiency. But it also means you need a much more powerful battery to start up the secondary power systems.
IIRC, this is the exact reason why Airbus opted that of that technology for its upcoming dreamliner direct competitor, even if the possible reductions in weight and size were very compelling.
The Tesla battery pack technology addresses precisely the avalanche-failure mode. Every node has a thermal cutout, so no fire is possible short of physically deforming the cell.
It's been a while and I'm no expert, but my impression from the reporting around the 787 problem -- and from some comments on HN -- is that Musk's technology was specifically engineered to be outside the envelope of these problems: Cell size, heat dissipation, cut outs, etc.
Boeing, by contrast, more or less attempted to introduce some... "buffering" into their extant design: (Additional) cell spacing -- as much as they could wedge in -- and maybe some other tweaks. Rather than really fixing the underlying problem.
I may be all wrong/wet, but that's the impression I was left with.
I think that if Boeing wants to salvage a good degree of near term profitability on this product line, they should begin talking to Musk. Actually, were it my call, even if the decision was to "fix" what they already had, I probably would have run a parallel examination and perhaps testing out of Musk's technology. Some millions on such an effort would be cheap "insurance", to bend that word -- no guarantee of a workable alternative, but a decent bet. And, even if you fix what you have, you now have a redundancy... and maybe you've learned something, along the way.
In short, Boeing needs to lop the top off a few internal egos, perhaps. Just speculation on my part, but...
What do you think of LiFePO4 chemistry? The energy density is not so great, per-volume it is on par with NiMH. But they can provide lots of current and have a very flat discharge curve and are pretty much fireproof. Oh, and have much better cycle count and thermal stability than LiPo. It is a very promising chemistry.
From what I understand, locking up the Dreamliner batteries in a fireproof container is one of the steps they took to limit the severity of any potential future issues. The idea being that the batteries don't have to be perfectly safe if they can't take down the airplane.
Yeah, I'm leaning heavily towards this not actually being the batteries this time. In a way, that is kind of worse I think since it means they are not dealing with a known problem.
heavily electrified big vessel? like submarine or a ship. At least at the start, there were a lot of fires from wires and other pieces of electric equipment. And you can't just go with extra thick wires in the plane design.
From that video, if it was the batteries, someone moved them, and then set them on fire.
The batteries are nowhere near there. They're low on the plane, one set near the wings and one forward. That's maybe the APU, though more forward than I'd expect. Looks more like the cabin, maybe even the "flight attendant rest area" which is above the cabin.
My guess: random electrical. Or someone smoking carelessly.
Here is a thing. The mishap rate (accidents and serious incidents) of a commercial airliner, including the 787 is magnitudes below that of driving to the airport.
The reason the 787 was grounded previously, was that Boeing and the FAA determined that the battery pack was insufficiently protected. The protection of the battery pack has been upgraded, both physically and electrically.
RC batteries and their charging systems are not nearly as thoroughly designed or protected as aircraft batteries.
Let's not exaggerate. Per mile, airliners are roughly 10x (one order of magnitude) safer than cars. It is generally true that driving to/from the airport is the most dangerous part of an airline trip, but not by a particularly large factor.
"safety/danger per-mile" is not a very reasonable metric in many cases though.
E.g., for airplanes, most of the danger is in takeoff/landing, and whether you fly 300 miles or 10000 miles in between is much less significant. For cars, long distances often involve highway driving, which is safer, and driving to the airport probably(!) doesn't involve alcohol, which is a much bigger factor in overall accident rates... etc etc.
What would be more interesting to know is the risk per flight, and compare that with a risk estimate for a given airport drive...
[In my case I just take the train to the airport from my house, and the lines I use have fatality rates which are essentially negligible (1-2 per century). I suppose I'm mostly likely to get hit by a car on the walk to the station... or just slip on the steps. >< ]
I had never thought about it like that, but that's a great point. The overall risk of dying on a 500-mile airplane trip is probably almost the same as the risk of dying on a 5,000-mile trip. Risk evaluation is hard.
Citation? We know of 3 fire mishaps on the 787, across 66 planes over 1.5 years of service. That's 1 mishap per 33 service years, which doesn't sound 'magnitudes' better than a typical road vehicle. Perhaps you have better statistics?
Fire departments responded to 194k vehicle fires per year in the US, across 254M vehicles. That's one serious fire every 1310 car-years. Thus, 787s catch fire 40 times more often than cars.
Just to add to this - I have been involved in 2x car fires. 1 had the fire brigade involved (car completely burnt out), the other 1 we dealt with and all was well (welding some rust at a friends and the boot/trunk caught fire). I image there aren't many plane fires where authorities aren't notified, but this is a guess. Does anyone have any knowledge of aircraft incident reporting rates?
120,000+ people killed in 4 years on the roads, and how many seriously injured? The Asiana crash was the first major accident in the US in 4 years. The 787 incidents have been non-fatal, and nobody seriously hurt.
I'm not sure the exact statistics, but how often does somebody crash their car on the way to the airport, including fender-bender accidents? A lot more than airliners sustain substantial damage.
Even though fire in an aircraft is a critical emergency, these 787 mishaps were only "incidents" and didn't cause substantial damage to the aircraft involved.
Doesn't a car typically break down at least once every 33 service years or so, too? For that matter, I wouldn't be too surprised if 1/33 cars crashed each year (or significantly higher than that), so I think that the 787 is still looking pretty reliable in comparison. (maybe not in comparison to a 747, but that's a different argument)
I'd like to nitpick that "magnitudes" depends on your base. Base 10 is commonly used, though base e is probably more accurate. So if 33/e^2 == 4.46 incidents happen, I'd probably call it "magnitudes".
I am always surprised by the way car versus plane safety come up. I know how to drive a car, and have control over the situation when I do. This doesn't mean its completely safe, but I can help control the risks. In a plane (or train, or taxi or any mode of transport where I am a passenger) I have almost no control and I expect the transport provider to provide me with transport that is safe. Very safe.
Of course in many cases, car drivers feel like they have much more control over things than they actually do.
Being in a commercial airliner on the other hand, actually emphasizes the feeling of helplessness: you're strapped in a seat and have little room to move, you cannot effect what's happening, you can't even see who's in control, or often even see your surroundings... it's like being trapped in a box tumbling down the mountainside.
I've found that even just unstrapping my seatbelt and taking a quick walk down the aisle on an airplane actually makes me feel safer, even though of course the reverse is actually true. This makes some sense given that human instincts probably value having the ability to flee or react in a dangerous situation but those instincts still exist even in situations where they make no sense.
Thus the perception many people have that they're "safer" in a car than on an airliner, even though the reality is that the odds are very much reversed. Large disparities between perception and reality can of course lead to people making very poor choices...
We should minimize the risks as much as possible. For RC aircraft (better chargers?), and definitely use fully optimized charging systems in electric vehicles, and for 787s.
How come we rarely hear about fires in Lithium-based battery packs in consumer electronics? I mean, it surely happens occasionally, but it has to be a very rare occurence.
~10 years ago it was frequent with some laptop producers (Dell comes to mind). I guess it comes down to quality of production - ie. don't try to be cheap bastard - and maturity of technology.
Both of this aspects are supposedly against the Boeing as the development (and thus certification, etc..) was done many years ago, thus they are behind technologically (like any other system/technology in aviation which have direct consumer counterparts) and Boeing is being run by MBAs which means they would typically be trying to be cheap bastards ("bottom line").
The iPhone 5 uses a lithium polymer battery, and has sold millions; are there reports of safety malfunctions with the batteries in that device? Or is they battery safer by virtue of design or lower power?
I've heard cases of iPhone 5s spontaneously combusting in china. I wasn't sure if they were just fakes or if there was a real issue with the battery. Thankfully, they are kind of small and the accidents are not fatal, yet.
> I know millions of Lithium-based batteries are in use
You're wrong. Billions are in use. Billions of laptops, tablets, cell phones, smoke detectors, and other assorted gadgets are sold every year with these batteries. There are several within easy reach of my chair right now. I bet if I tracked down every lithium battery in my house, there would be at least a couple dozen accumulated over the years.
Just today, the New York Times said[1] "The Consumer Product Safety Commission has documented more than 350 fires involving lithium-ion batteries since March 2012".
350 fires out of the hundreds of millions of these batteries that must make their way into Americans' hands every year.
Perspective: In 2012, 28 Americans died from lightning strikes[2] (a below-average year, at that). In 2010, over 4,000 pedestrians died being in the US[3]. You don't want to know how many people died in 2011 to tornados.
I can't think of a single death that has been attributed to a lithium-ion battery fire, and only a handful of injuries.
If lithium batteries aren't safe, I don't know what is. Boeing has done something deeply wrong.
So what? The number of Li ion batteries doesn't really mean much.
By analogy: there are at least something 10^34 (10 million billion billion billion) uranium atoms on earth, and they weren't explosive for billions of years, yet somehow they seem to be potentially quite explosive since 1945.
It's a question of critical mass. A small lithium ion battery may be safe, a large one may not be. Just because there are billions of tiny lithium ion batteries on the world that are relatively harmless doesn't mean large ones aren't - the energy capacity rises in the cube of the size, but heat dissipation only quadratically.
In short, because it works safely in the small doesn't mean it's easy to implement safely at scale.
They used Lithium batteries because they are light, small and powerful.
Can Lithium batteries fail? Yes. But keep in mind that flying along with you and the batteries is many tons of flammable fuel, wrapped in aluminum and carbon fiber, which are also flammable, with miles of high-powered electrical lines running through it all. And we know from TWA800 that those are enough, by themselves, to bring down an airliner.
But we also know that that happens very rarely, because engineers can mitigate dangers if they are aware of them.
This incident is clearly not directly related to the previous electronics bay battery issue. It could be indirectly related -- an electrical short in the cabin, for example, which would be fed by the batteries.
