Another big recall 44

[enginesrus]

https://www.cnn.com/2023/09/06/business/air-bag-recall/index.html

Will they ever learn a tin can with no pressure relief system, is just not the proper pressure vessel.

 

[JohnRBaker]

It appears that the auto manufactures are fighting the latest recall order:

Carmakers Fight Massive Air Bag Recall That May Cost $10 Billion

https://www.bnnbloomberg.ca/carmake...bag-recall-that-may-cost-10-billion-1.2014146

John R. Baker, P.E. (ret)
Irvine, CA
Siemens PLM:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[dik]

Thanks John... I didn't know that Takata went into bankruptcy. It's a pretty sad situation where something is intended to act in an emergency occasionally goes amiss. The net benefit is really there.

IRS... do you have any numbers that show the change in fatalities after airbags were generally implemented?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Sparweb]

The agency said in May it had identified at least seven cases of ruptured air bags that led to injuries, including two deaths, dating from 2009 to as recently as this past March. Regulators believe welding performed in the manufacturing of ARC inflators may have left debris inside the part. During a crash, gas produced by ignition is supposed to fill up the bag. But when the channel is clogged, it can cause excess pressure to build up inside the inflator and potentially spray metal fragments.

This is helpful information. The time period from 2009 to today covers 14 years. In this time there have been "at least seven" ruptures reported to NTHSA or IIHS. During these 14 years there have been over 500,000 fatalities in vehicle accidents in the USA (estimated somewhat because the IIHS data ends at 2021). Actually only 350,000 fatalities in passenger vehicle accidents (to omit pedestrians, bicycles, motorcycles, etc.) These deaths occurred in about 480,000 passenger vehicles involved in fatal accidents, as reported to NTHSA or IIHS.

This is where we have to be careful... There is no information that the ruptured airbag failures were the primary cause of injury or death. Technically it's not the same thing as being associated with the injuries or deaths, nor is it clear if these accidents were likely fatal without the airbag or not. Let's sweep aside the uncertainty by just deciding that the airbag failures are independent of the nature of the accidents. It makes the math easier but it may over-estimate the criticality of each rupture.

7/350,000 = 2.0 * 10^-5 ruptures per fatality
7/480,000 = 1.4 * 10^-5 ruptures per passenger vehicle in a fatal accident

In each fatal accident in a passenger vehicle, there are numerous air bags involved. Almost always one for the driver, and for any passengers of sufficient weight to activate the airbag. There are also side-curtain and even belt bags, now, depending on the type of car. So for each automobile in a fatal accident there could be 1 or 6 air bags activated. Let's assume only the driver's bags are activated and the typical car has only 2 per front seat.

7/960,000 = 0.7 * 10^-5 ruptures per airbag activated in a fatal accident

I wanted to put this in context because this failure rate is technically a bit high for a safety/protective system. Maybe that's what has the NHTSA all up in arms. A more acceptable number is one failure per million activations or 1*10^6. But it's not actually far off the target. And I made some conservative assumptions.

What really shakes me, though, is the sheer HUGE number of fatalities on US roads. 40 Thousand deaths per year! Half of these fatalities were people NOT wearing a seat belt. Not even wearing a seat belt!

Any effort from the NHTSA or IIHS or US government that can move the seat belt usage by one thousandth of one percent will have a corresponding effect on the US fatality rate that would overwhelm any success they have on recalling airbag inflators.

https://www.iihs.org/topics/fatality-statistics/detail/yearly-snapshot

 

[dik]

Aren't they two different conditions. One safety situation where by your actions you are requiring a 'safe experience' and the other as a result of a violent activity, it is protecting you to provide a safe experience. One is passive and the other is active.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[IRstuff]

A more acceptable number is one failure per million activations or 1*10^6.

However, it looks like you were comparing deployments in fatal accidents, as opposed to total deployments, since the 7 deployment ostensibly would have not resulted in fatalities.

https://www.scienceservingsociety.com/ts/text/ch12.htm

suggests 6.6 deployments per 1000 vehicles per year, so 14 years times the cited 74 million airbag equipped cars divided by 1000 times 6.6 = 6.8 million deployments. However, the same site points out that there are literally thousands of deaths where airbags didn't deploy.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[JohnRBaker]

However, the same site points out that there are literally thousands of deaths where airbags didn't deploy.

Could the welding problem, leaving debris behind, also be related to the incidents where airbags did NOT deploy?

John R. Baker, P.E. (ret)
Irvine, CA
Siemens PLM:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[3DDave]

In the ARC design there is a single exhaust port for the gases to escape the gas generator. The generator can do only two things - either produce a constant stream of gas or, if the port is plugged, release all the contents as gas instantly. The theory is that weld spatter would move to plug the port, partially or completely, causing rupture. Either condition will deploy the airbag.

Deaths where the conditions were not met to deploy the airbag but were still fatal can certainly happen - partial ejection during roll-over, for example.

It seems unlikely that debris would interfere with the ignitor.

 

[Sparweb]

Dik, IRStuff,
I could indeed be comparing apples to oranges, by putting different fatality rates together. But the REAL point was (I hope) made by the other comparison, of how vanishingly few flaws there are to the enormous number of accidents that aren't related to these flaws in any way. So Dik, your point may double my estimate of reliability, and IRStuff's point may make it half, but those corrections are peanuts compared to 6 orders of magnitude more deaths in the annual fatal accident rate.

 

[thebard3]

The world has changed. When I was a liitle kid, we climbed around on our parents laps in the front seat where there were no seat belts and where your head would now contact an airbag was then a big chunk of steel called the dashboard. Now somehow we have become so risk averse that we expect the developments and investments of others to provide absolute and total protection from our own carelessness and stupidity. Somehow, now regulations have evolved to support that premise. Driving around in your car has an inherent danger and always will. There is never a statistical probability of zero for consequences of our actions. Carmakers should only be expected to supply what is determined to be the best available technology that the consumers will accept (pay for), and that evolves over time.. It's easy for lawmakers to create regulation. It's usually much more difficult to implement.

Brad Waybright

The more you know, the more you know you don't know.

 

[enginesrus]

SwinnyGG, Its so nice when we can just debate the technical on this site and not spend to much space picking on others.
I don't exit the thread, It just sometimes gets lost between other threads, and I have other things in life to do as well.
Someone mentions how the air bag is the pressure relief device. Then there is no reason for an inflator to fragment then.
Most seem to think it is a non issue. I'm glad they don't deal with aircraft problems this same way.

 

[Sparweb]

Unintended consequences.
When the public does not trust safety equipment (I am old enough to remember the hoopla about introducing airbags in the '80s) they don't use it, they don't accept it, they even try to defeat it, sometimes. This happened often in the 1980's where people would deactivate the airbags, especially when they heard about children being killed by them. The NHTSA tracked dozens of children killed per year (not kidding!) for about a decade before the auto OEM's were able to implement seat pressure sensors.

Tampering with safety equipment like this won't go away. Today, if the bad press about bad airbags takes hold, and reaches a wide audience, how many more car owners will disable the airbags in their own cars? Out of a million vehicles maybe only a handful. But I'm sure that some people will take it on themselves to disable the airbag because they are convinced it is dangerous. This happened in the 1980's for the same reason. Out of all of the vehicles across North America it may not be common, but how common does it have to be for this to have its own safety impact?

Rate of airbag ruptures per year in the US = 0.5 deaths
Number of motor vehicle accident fatalities in the US, 2021 = 26,585 deaths
Number of automobiles in the US, 2021 = 282,366,290

Rate of of motor vehicle accident fatalities in the US, 2021 = 26,585 / 282,366,290
= 9.42*10^-5 fatalities per vehicle

Tampering rate needed, to increase statistical likelihood of 1 death per year:
1 death / 9.41*10-5 death / vehicle = 10,621 vehicles

So this isn't proof of anything, but given that tampering with airbags was widespread when they were introduced, and deactivation of airbags is still a common thing, it should not be a surprise that they are still tampered with. I don't have statistics on airbag deactivation today, but I'm not going out on a limb to say that it is more than Zero. How much worse will it get due to bad news about "unsafe airbags" as promulgated by the NTHSA now?

Do I believe tampering will be so bad that 10,000 more vehicles will have their airbags disabled by owners?
Maybe. If so, then the number of lives lost could exceed the number of lives saved by airbags that do work.

Are legitimately deactivated airbags common? Yes, absolutely!

https://tc.canada.ca/en/road-transportation/publications/air-bag-deactivation/air-bag-decision

https://www.nhtsa.gov/equipment/air-bags#the-topic-on-off-switch

Is it a common after-market mod? Yes, indeed:

https://www.dormanproducts.com/p-67295-924-900.aspx

https://duckduckgo.com/?t=ftsa&q=Air+Bag+Deactivation+switch&atb=v272-1&ia=web

 

[dik]

I think we are getting caught up in the numbers, a bit. When you have an inherently dangerous mechanism that activates a device the saves far many more lives than it may 'take' there should be some latitude to accommodate the dangerous nature of the rapid deployment. These lives may have been lost anyway.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

Disabling the devices would be akin to not wearing a seatbelt or not wearing a motorcycle helmet... do so at your peril.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[enginesrus]

But most importantly the actuation device could be made fool proof so as not to cause an injury.
Its a huge life changing injury that would worry some of us the most.
So the question is why no huge effort to safe a fy, the inflator?
I would like to know how in Gen 1 LS400's they got away with using a machined aluminum case for the inflator, that was about 1/4 inch thick?
Safer propellant for 1 I'm sure.

IRstuff, just point me in the direction to deliver my improved design to? Per your second post in this thread.

 

[3DDave]

Try

http://www.arcautomotive.com/careers.html

 

[dik]

The old expression it's difficult to make things foolproof... they can be so ingenious... comes to mind. Even with careful control of an explosion the results are mostly comparable. The devices should use the criteria that they are 'mostly' safe and will provide protection most of the time or at least reduce injuries most of the time. The criteria should be a little stronger than this so that they nearly always work as intended. I don't know how to stipulate that in numbers. Air bags greatly enhance your likelihood of living, but maybe not all the time. It's not that the airbag 'killed' you but that you likely would have died anyway.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[IRstuff]

But most importantly the actuation device could be made fool proof so as not to cause an injury.
Its a huge life changing injury that would worry some of us the most.

Aside from fools, electromechanical objects cannot be made failure proof, within practical limits. We do not demand cars themselves to be collision or failure proof, only that they meet some level of acceptably low failure rates. Nor do we expect airbags to protect us 100%. That's just a reality of life; just like we can't even guarantee that any microprocessor that leaves the factory is 100% functional, because it would take nonsensical amounts of testing and time to verify that each and every transistor among the billions on microprocessor are all 100% functional; we gave up on that nearly 40 years ago on older microprocessors. We don't even bother running power on self test anymore because the time involved would be ludicrously long, and no one wants to wait 10 minutes to use their laptops. I've been involved in reliability analyses of flight data recorders, and there's no expectation of 100% reliability and the ppm-level failure rate is the normally acceptable failure rate, simple because the cost to do better would approach being prohibitively expensive.

And, no one is willing to pay to even do the testing to prove such reliability statistics; someone would basically have to spend hundreds of millions of dollars to run sufficient testing to demonstrate such reliability levels, which would jacke the cost of safety equipment to be an unacceptably large portion of the cost of a car.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Alistair_Heaton]

How about you get rid of the issue with the fact that US airbags are different to the rest of the world because they have to be able to work without the seat occupant wearing a belt.

This leads them to have to deploy faster and with more force.

 

[dik]

didn't know that... thanks Alistair

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

Can fools be made failure proof? Just curious...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik