Takata’s Air Bag Crisis | Hacker News Reader

Takata’s Air Bag Crisis

bloomberg.com

81 pointsusaphp69 comments

There is little doubt these bags are dangerous and pose a known threat to passenger safety. But to call these car bombs is irresponsible. They are not purposely armed with the intent to inflict damage to people. Yes, they have design flaws which have resulted in deadly injuries. But while technically these are explosives to aid in deploying the safety mechanism, they are not bombs as most people understand things.

This is as unjust as calling people who get arrested "disappeared" with all the connotations that word has (i.e. secret summary executions). This has the same approach, a bombastic approach to headline writing. For shame!

The implication here is corporate malfeasance which is exactly what was displayed. How many design reviews have you been present where an engineer said,

    “If we go forward with this, somebody will be killed”

At that point, I'd lay my badge on the table and resign then and there.

I think once they've degraded a bomb is apt. Not in the hollywood exploding car sense, but like a more realistic anti-personnel explosive complete with shrapnel. In most real bombs its the shrapnel that kills not the blast itself.

When I read that they used ammonium nitrate first thing I thought about was: "That's bomb-material, how can you put that into a running vehicle?". One of the worst road-side accidents that happened in my country was this: https://en.wikipedia.org/wiki/Mih%C4%83ile%C8%99ti_explosion

> The Mihăileşti explosion occurred on 24 May 2004 in the village of Mihăileşti, Buzău County, Romania. A truck loaded with ammonium nitrate rolled over and caught fire, and one hour later exploded killing at least 18 people and another 13 were wounded

Notice the "one hour later" part. They are indeed ticking time bombs.

> That's bomb-material

how else do you suggest they produce enough gas within milliseconds to inflate two or more airbags ?

As I understood from the article ammonium nitrate is way more unstable compared to the other solutions. Apart from the Oklahoma terror attack and the case I mentioned there is also this (https://en.wikipedia.org/wiki/2011_Norway_attacks#Breivik_Ge...):

> The place of business was given as Åmot in Hedmark. On 4 May 2011 Breivik purchased six tonnes (13,000 lb) of fertilizer through Geofarm at Felleskjøpet, three tonnes (6,600 lb) of ammonium nitrate and three tonnes of calcium ammonium nitrate. According to neighbours, all the fertilizer was stored in his barn.[47] After conducting a reconstruction of the bomb with equivalent amount of fertilizer on the farm in Åmot, police and bomb experts concluded that the bomb had been 950 kilograms,[48] about the same size as the one used in the 2002 Bali bombings.

All I'm saying is that they could have chosen another explosive-like material, not one which is known to be unstable and which has been used by terrorists to make bombs exactly for this reason.

(Note, all this is based reading, explosives are something I've thoroughly avoided, I like my fingers.)

Nope, ammonium nitrate is a secondary explosive, and as far as I know a rather insensitive one, although how insensitive would depend on what fuel you mix it with. I don't recall if it was mentioned if Takata is mixing it with one. By itself, it's quite insensitive, but in large quantities in worse case conditions has resulted in some of the worst explosives disasters: https://en.wikipedia.org/wiki/Ammonium_nitrate_disasters Note the Texas City one was in theory pure ammonium nitrate (although I wonder if it was pure enough...), many of the others had other stuff contributing, or starting the disaster.

Don't know about sodium azide, but lead azide has been used as a detonator, and therefore a primary explosive as far as I know, in non-corrosive firearms primers for some time (although I think lead styphnate is more popular).

Ammonium nitrate is a preferred explosive component for many classes of users, military, civilian, and terrorist, simply because it's very very cheap, it's also used as fertilizer. Which of course then makes it relatively accessible to the latter class of users, but not in the ideal form, the more surface area, the higher the boom.

Which circles back to the Takata problem, they haven't been able to keep it from turning into a powder, which has the max surface area and turns it into an explosive that turns the air bag into a true, metal fragment and splinter producing bomb.

The difference is the speed of gas production. A bomb has an uncontrolled rate, and so is lethal by intent and design. The inflator in an airbag should produce gas rapidly, but in a controlled manner - more like a rocket engine. The design of the Takata airbags meant that the tablets of fuel would break down into a powder over time, vastly increasing the surface area that was available to burn and thus gas production became uncontrolled.

Reminds me of http://www.dhmo.org/facts.html

As mentioned in the article, which I hope you have read:

> But ammonium nitrate had a critical flaw that he says led other air bag makers to give up on it: Ammonium nitrate has five phases of varying density that make it hard to keep stable over time. A propellant made with ammonium nitrate would swell and shrink with temperature changes, and eventually the tablet would break down into powder.

It seems like ammonium nitrate is more unstable compared to the other solutions (this is why it makes for good bomb material, see the Oklahoma terror attack or the Breivik terror attack in Norway), so it was a very bad decision to use it. But if you want to hit your straw-men then go ahead.

> It seems like ammonium nitrate is more unstable

Bloomberg, I think, used the word 'stable' incorrectly in the excerpt. It's not unstable — i.e., it doesn't just go off without being ignited — it's unpredictable. I.e., what should be a relatively slow explosion instead is a very fast one, increasing pressure & thus brisance, leading to shrapnel.

I didn't RTFA, did it happen to quote total recorded deployments of these bags vs noteworthy injury vs death?

13/100+/60,000,000-100,000,000 is the known tally. Not sure about injuries or if such numbers could be reliable as it could be hard to trace source of injury conclusively (as there may be many parts in motion or contact during a collision induced deployment.

> [Shigeshisa Takada] didn’t mention that Takata had tried to fix the problem by changing the propellant formula in 2008.

My Acura TL is affected by this airbag recall, and it's a fairly recent 2012 model too. Whatever changed circa 2008 apparently didn't work.

> NHTSA says those companies are making 70 percent of the replacement inflators.

When I took it to the dealership for servicing about a week ago, I also inquiried about the airbag issue. According to service rep, it's a quick 30 minute fix assuming parts on hand...problem is the 30-45 day lead time for parts, which blew my mind away considering the nature of the issue in a state well known for its warm weather and humidity. Considering the logistic predicament, I wonder how quality will be affected given the number of 3rd-party players manufacturing a complex replacement component for a proprietary airbag system.

The 2008 fix, according to the article, started in Europe and was also used in China.

1998 article on the Paresh Khandhadia, the Takata engineer

http://www.autonews.com/article/19980223/ANA/802230779/slow-...

So, he basically created a very safe airbag inflator (tetrazole) and later screwed everything up by going with ammonium nitrate? WTF!

A great longform article, one of the best indepth overviews of a company's culture and current controversies that I've read in awhile...and a pretty egregious controversy at that.

Check your car by VIN here: https://vinrcl.safercar.gov/vin/

I think this story also shows the problems with patents. If Takata was allowed to use the mixture that was "invented" by other airbag makers, would they have used their dangerous ammonium nitrate mixture?

It also shows the dangers of the drive for lower costs in companies like this. The current system is doomed to reintroduce the same problems all over again, regulation and more control will not change this.

The other side of that question is: if other airbag makers weren't able to patent the better mixture, would they ever have bothered to invent it?

I'm not implying that the answer is necessarily "no," but the mere fact that patents discouraged the use of a safer mixture in one case doesn't tell the whole story.

> If Takata was allowed to use the mixture that was "invented" by other airbag makers

It doesn't invalidate the question as a whole, but they would have been allowed to if they licensed the solution but that may have cost them more than developing their own (now known to be badly floored) solution. An active choice was made there.

If Takata was allowed to use the mixture that was "invented" by other airbag makers, would they invest money into inventing them in the first place?

The article mentions Honda's efforts to track down owners of cars with defective airbags. Why can't state motor vehicle registries cooperate in recalls like this?

I have an 11 year old Honda subject to the recall. I'm the third owner and I bought it from a dealer with no Honda affiliation. Honda has been bombarding me with notices, suggesting that the state _has_ been helping.

What I'm more annoyed about is the UX. Honda is getting press puff pieces published about what an awesome job they're doing up to and including rental cars for people who are concerned while parts are on backorder. Meanwhile my local dealership told me that they'd drop me off/pick me up in their shuttle as long as I wasn't more than 2 miles away.

User experience is going to vary widely because dealerships vary widely.

The dealership where I bought my Honda gave me a loaner while my vehicle was repaired. I dropped my car off on the way to work, and picked up the loaner. The work was done by the time I got off work. The whole experience was painless.

A federal agency constructively coordinating with multiple state agencies...I believe that's considered an emerging paradigm shift.

EDIT: To be sure, Honda isn't the only major vehicle manufacturer involved [1]. Observing that this isn't an explicit state issue and someone's got to foot the bill, this would almost certainly require some federal agency (e.g. NHTSA) compelling applicable states to act, a bureaucratic process that never goes smoothly in practice.

[1] http://www.airbagrecall.com/home/

> Takada has only $520 million on hand and is worth about $340 million

Does this mean they have $520m in cash, but the stock valve is less than their bank account?

Sounds like it. Take into account the impending fine, all the lawsuits in many countries, the family control of the firm... Takata at this point could well be worth nothing at all.

They have other product lines (the article mentions their seat belt business), but my guess would be that the airbag liabilities would vastly exceed the income & valuation of those.

Yeah, the uncertainties about their future must weigh heavily on their value. If they fail to meet commitments set by the US they're liable for another 130mm in fines. That doesn't address the possibility of future lawsuits either.

They probably have debt aswell.

Seems very likely considering Japanese interest rates are near zero.

Very interesting read, and makes me recall back to being a kid and setting off airbags in the back yard. You could immediately tell the difference between the older propellant and the newer ones, though I wasn't aware of how poisonous it was for us.

My Subaru dealership can only order 5 airbags per day. I was a couple hundred deep in the list. They said it takes 4 hours to do one, but Subaru claims 2.

I'm going to be waiting for quite a while.

2 hours to pull it into the bay, 45 minutes to contemplate life, 15 minutes to swap the air bag, and 1 hour to day dream about an FA20 WRX STI. It's seriously so simple to swap, they are lying.

my 2005 outback has to have the entire dash removed to do it.

Passenger side I assume, that would definitely take a long time.

Correct!

http://hondaairbaginfo.com/news-updates/statement-by-america...

"Automobile models and model years included in the first stage of this recent expansion of the Takata non-desiccated passenger front inflator recall"

Doesn't make me feel warm and fuzzy as a 2006 Odyssey owner.

Yep, talking to myself:

http://www.nhtsa.gov/About+NHTSA/Press+Releases/nhtsa-expand...

"The five recall phases are based on prioritization of risk, determined by the age of the inflators and exposure to high humidity and fluctuating high temperatures that accelerate the degradation of the chemical propellant."

So I guess I'm a phase 2 or 3 kind of owner. Meanwhile, sitting at the vaunted 10 year marker in a hot 'n humid climate...

In your situation I'd probably just yank the fuse until the replacement occurs. On that car it looks like it's in the interior fusebox on the driver's side, fuse #22, "IG SRS". You won't have any airbags in an accident with it pulled, but you also won't run the risk of getting the equivalent of a shotgun blast of metal shards to the face.

At that point, I think that personally I would have it taken out.

As I understand it, the problem is that the explosive becomes more powerful over time, too powerful. Here's a wild idea --- how about "pre-aging" the explosive until it's reached basically its maximum "strength", and then filling the airbags with a smaller amount to provide the same equivalent force as before? It could save them even more... and they could spend some of the savings on making the casings stronger and thus less likely to fragment.

It is not the force that changes, it is the rate of gas generation (burn rate). You need to generate airbag gasses fast enough to inflate before your face hits the airbag, but slow enough not to tear itself apart (especially the metal forming the gas generator).

As the ammonium nitrate based gas generator mix ages, becomes less stable in that it burns faster, generates its gasses faster, and crosses the threshold between gas generation and explosion (RUD).

Ref: https://en.wikipedia.org/wiki/Gas_generator

Does it become less "stable", or just much higher in speed of detonation? I don't remember reading any accounts of these airbags going off when they shouldn't have, just bad results once the system triggered them.

Stability here is being used in the chemical sense, as in "how fast it oxidizes". https://en.wikipedia.org/wiki/Chemical_stability

A gas generator (airbag inflater) does not detonate, it generates a large volume of gas very rapidly but at a designed (predictable) rate - the faulty airbags are detonating.

Spontaneous ("accidental") deployment would be the ultimate bad instability, but I don't believe this was a major problem (yet). IIRC, there was one reference to an "accidental" deployment which could be instability-related or a fault elsewhere, e.g. in the sensor/trigger circuitry.

The instability that is causing the major problems is causing faster combustion than designed.

Ammonium nitrate is at least non-toxic. You can sprinkle it on your soup. The exhaust gases are likewise harmless, and even have enough oxygen to keep you alive.

Sodium azide is deadly both ways. Unreacted azide is nasty stuff. Sodium makes lye when it gets in your eyes or lungs, and is also an extreme fire hazard. There is no oxygen.

the article mentions 13 deaths of which 10 in the US. This seems a very odd distribution, can someone explain it?

I can't explain it, but I have a couple of qualified guesses:

a) Takata air bags, or more precisely - Takata airbags using the faulty initiators - are disproportionally allocated to the US market for some reason or the other. Perhaps other Takata plants supplying other markets use a slightly different explosive compound than the one being made at Moses Lake?*

b) Perhaps authorities in other countries haven't investigated past crashes and concluded that the airbag caused the fatality, rather than the crash itself, hence they cannot claim that the airbag was to blame for any fatality.

c) Perhaps they do have confirmed fatalities b/c of a Takata airbag, but haven't reported it to whatever authority provided the numbers for the article (yet, anyway).

*) Bingo; found this as I read the article: "He says he hired a propellant specialist to help develop a more stable formula using guanidine nitrate, and since about 2008, Takata in Europe has sold air bags using that. He says Takata’s China team also adopted the formula."

I think tmd83 might be on to something (as well as m_mueller), the NHTSA decided in the ... '70s I think to make the tradeoff to maximally protect those who don't wear belt.

But environmental conditions might also be a factor, the higher the heat and humidity, the faster the ammonium nitrate pellets turn into very dangerous powder. And that's being used to target the limited supply of replacement parts in the US. How does Japan's population distribution and climate compare to the US?

The (very good, worthy of a read) article deals with this. It states that the european subsidiary of Takata developed a propellant based on guanidine nitrate because of concerns with the safety of ammonium nitrate, and have used that since 2008. It also states that their Chinese subsidiary used that one.

Here's my guess: The US and Japan are the largest markets for Japanese cars (I think). Japanese who drive Japanese cars tend to switch to the latest model fairly often, at least more often than Americans. At least I'm seeing way more old cars when visiting the US.

Down the thread in the quora answer there is mention of US airbags deploying at higher pressure to try safeguarding people who aren't wearing seat belts. Could that potentially make the US versions more failure prone and unsafe?

Would it be better to deactivate my airbag until it can be replaced, or run the risk of having this issue pop out, assuming my car is affected?

If it is worth anything I drive like a maniac.

Step 1 would be to stop driving like a maniac. Otherwise since there have been 13 deaths for millions of air bags, probably still safer to leave it activated. The chance of serious injury from your head striking the steering column is still significantly larger than shrapnel from a defective air bag injuring you.

My 2007 CR-V is affected. Since I live in the northern part of the US, my airbag isn't a priority as they're doing southern states first.

I live in one of the driest states, and my air bags were swapped a year ago.

> Torture the data until it confesses

such a sad and tragic misunderstanding of the meaning of this phrase.

One of the early criticisms of air bags in passenger cars was that they were said to protect only people who aren't belted in, with nothing but potentially-injurious effects for anyone who is.

Does anyone know if that is still (or was ever) true?

I believe that was never true for vehicle occupants in the typical ranges of height, weight, and age.

Failure to use seat-belts with earlier generations of airbags tended to lead to unbelted occupants "submarining" under the opening airbag, leading to lower body injuries, although the "submarining" jargon now appears to be applied to situation where a lap belt is wore higher on the abdomen rather than low on the hip bones, with the higher position tending to lead to abdominal injuries.

With newer generations of airbag design and other tech such as active seat belt pre-tensioners, I would be extremely surprised to find any recommendations against wearing seatbelts, except perhaps for the 1 in 100,000 cases - near-term pregnant women or persons with recent abdominal surgery.

Here's a detailed review (from the mid-90's) http://www.nhtsa.gov/Laws+&+Regulations/Air+Bags/Third+Repor...

and more recent reports http://www.nhtsa.gov/Driving+Safety/Research+&+Evaluation/Be...

http://www.cdc.gov/motorvehiclesafety/seatbelts/facts.html

Wearing a seat belt is always better. In the US they do adjust airbags to work for people not wearing them. Making the experience worse for people wearing a seat belt. https://www.quora.com/Is-it-true-that-airbags-in-a-car-dont-...

Air bags work far better when you're belted in. The seat belt slows the occupant down so the air bag has more time to inflate and has less of a blow to cushion.

Seat belts don't necessarily keep you from wanging your head on the steering column or dashboard. Their behavior while resisting the dozens of gees of a crash isn't the same as when you just tug on them normally. They stretch out and dissipate energy but you can still be moving at a good clip when you get to the hard bits up front, unless an air bag gets in the way.

There's a lot of misinformation about air bags out there. For reasons I don't really understand, some people find them to be extremely objectionable, and this results in a great deal of material which greatly exaggerates their dangers and downplays their benefits.

I know someone who was hit head on. The airbag deployed, but she also had a bruise on her shoulder from the seatbelt. I think they work well together. I advocate both.

I think part of the confusion might be in the US at least airbags are part of the "passive protection" which is designed to protect you if you do nothing (before airbags where afordable there were all manner of pretty bad automatic shoulder belts for cars provided the mandated protection.)

https://www.google.com/?gws_rd=ssl#q=automatic+seatbelts

There are many other good points being made in this subthread, the one I'd like to make is that if a driver is wearing a seat belt as he should, if he doesn't have an air bag and has a >= 35 mph collision, his head is going to hit the steering wheel, with worse and worse effects as the speed increases.

And without other equipment that's not practical to put in civilian vehicles, that sort of give in the seat belt system is required to avoid other bad effects like snapping necks.

It's all a carefully engineered system of tradeoffs, modulo Takata skipping the careful engineering step....

Reply on news.ycombinator.com