So what is the actual failure rate of these switches? Out of every 10,000 switches made, how many have failed? How does this compare to failure rate of switches made by other manufacturers? What actually fails, the switch itself or the Micky-Mouse column lock and rod assembly that activates the switch, usually mounted at the base of the steering column?
TBF, time before failure…Every mechanical part or device has one…Nothing lasts forever…55.000 people die in car wrecks every year…What else is new?
The failure mode is that the switch accidentally turns off because the detent spring preload is too low…ie it’s too easy to turn the key off by accidentally jostling it, etc.
No one knows the true number of “accidental shutoff” incidents because most of them don’t result in crashes, and therefore don’t make it into official statistics. But I’m guessing that number is quite a bit higher than the number of crashes. One woman who was leasing a Cobalt had three accidental shutoff incidents, none of which resulted in accidents, but were dangerous nonetheless. She called the car a “deathtrap” and returned it to the dealer even though they refused to refund her lease balance. There are probably many such stories.
Much like Nixon and Watergate, the coverup is worse than the crime. It sounds like GM did just what Ford did with the TFI module problems.
Apparently Ford was warned that the fail safe temperature on the modules was 257 degrees and the modules could fail around that point, Given the choice of re-engineering at a cost of millions or saying full speed ahead and take our chances, Ford chose the latter.
The ensuing lawsuit settlement cost them far more than re-engineering would have.
I agree, sounds like they made a conscious choice not to recall the vehicles with the defective switches and instead settle the inevitable lawsuits. In the Times article yesterday, there was a quote from a GM person involved in that decision saying there was no “business case” for recalling the switches.
That “business case” mentality that says, yeah, let’s keep on selling dangerous products because settling lawsuits from dead peoples’ families is cheaper than fixing them is what has got to change. And I think this case at GM will start to turn that mentality around. It’s never good business when consumers see you’d rather make a buck than make a safe product.
I think that current GM management saw what was happening to Toyota as a result of the unintended acceleration backlash and decided that it was better to get in front of the issue, even at this late date, than lag behind and end up with fines over $1-billion.
Just my 2 cents as usual and I’m not excusing GM for one second, but…
There’s not a car maker alive who hasn’t tried to cover something up.
Many other makes of cars randomly stall for any one of a variety of reasons.
While tragic, one of the deaths being cited is a 16 year old girl with the claim that she was killed because of the switch position and bags not going off. Some points to ponder.
Maybe that switch was turned in a last second panic.
That girl had a fight with her boyfriend at a party and left mad with foot on the floor. She had been drinking and was not wearing a seat belt when she smashed into a tree.
Some family members are appalled and express indignation that a car maker could do this or that and vow to hold them accountable while letting the world know about a Death Car.
A few years go by and that indignation they felt is shunted aside when offered a 6 or 7 figure settlement. They sign that confidentiality agreement in exchange for a hefty check and quietly go away with now fully muted indignation.
I’ll still reserve judgement but when you say GM, you are talking about thousands of people, some good some bad, and sometimes one wins over the other. Never been there? Your skin flint boss, with the lawyer and the bean counter, and you try to tell them to just do what’s right without a lot of hassle and be done with it. Its hard to justify to them and sometimes they just would rather put someone through the wringer first. My boss and I and the lawyer are in much better terms now that we’ve all retired but I still don’t like the bean counter.
FYI, GM is now recalling another 1.3M vehicles to fix “sudden loss of power steering” complaints on vehicles with electrically-assisted PS. Not related to the ignition switch issue, but it doesn’t inspire confidence.
One thing you have to remember though is that issues like failing intake gaskets are an engine design/gasket design issue that should have been caught by GM and they had tons of time to revise it on new vehicles–so the fault is entirely on them for this IMHO, just like Subaru’s head gasket problems and Toyota’s and Chrysler’s sludging issues with certain engines. Though at least Toyota and Chrysler seem to have either redesigned or phased out the affected motors–I believe some GM engines still have intake gasket problems.
Something like an ignition switch or power steering assist motor is not normally built in-house by GM or other manufacturers. A set of specs are issued, a (cheapest) supplier is sourced, and failure-mode testing is done by the supplier. Sure, GM’s quality control should have caught it before now, but really you could pin the problem on whatever top-tier supplier actually designed and provided the parts. (correct me if I’m wrong and someone knows if the part was designed/built in-house) It is however criminal (or at least criminally stupid) that problems like these were likely well known throughout the company but top management didn’t know or weren’t informed. So either someone is lying or management at that level is truly too high to see down, or people low on the totem pole work in an environment where they so fear reprisal that they will not report something like this.
What’s strange with the ignition switches is that the supplier, Delphi, TOLD GM that the ignition switch design did not meet GM’s specs for operating torque. And apparently GM ignored the report and had Delphi go ahead anyway and produce millions of substandard switches.
It’s mind-boggling. Your supplier says to you, Hey, these switches you want me to mass produce don’t actually meet your own specifications. And you say, that’s OK, just go ahead and build a few million anyway.???
Maybe the thinking was, “We’re in too much of a hurry to change the switch design now, let’s go into production and we’ll fix it later.” That might explain why the switch design was changed a few years later at a very low level, apparently without management awareness.
Mary Barra just faced the congressional committee and after apologizing told them that GM will make things right. Presumably that also meant they will not stonewall their responsibility by claiming the bankruptcy no longer made them responsible.
The electric steering gear is next on the recall list.
Its really easy to armchair quarterback this problem. It is a comedy (tragedy) of errors but consider a few points. The torque spec was defined by GM to supplier Delphi, once a part of GM. The spec was not a do-or-die type spec. It was developed for the proper function and “feel”. That “feel” part is totally subjective, the function part may not have been developed with a 2 lb keyring in mind. The engineer that approved the deviation that allowed the low-torque parts, may have evaluated how the ignition “felt” on the test buck. All was well so acceptance was granted. The later change was likely made with ignorance of the field failures. The torque was closer and the “feel” was good. If no complaints had yet surfaced, use the same part number with a revision. Standard industry and GM practice, no malice intended. I suggest this train of thought only because I worked for Delphi even when owned by GM. I have experienced how things get done at GM and their suppliers.
Now the history, Delphi filed bankruptcy in October 2005. The ignition switch division was to be sold off. No buyer could be found, GM bought them back. Then GM filed for bankruptcy. Do you think this upheaval could have caused things to slip through the cracks? Yeah, it could. NHTSA knew and they didn’t force a recall. What did they know about the complaints and what did they do about it?
Was there ANY consideration that GM went through bankruptcy and was now partly owned by the very government that chartered NHTSA? Did THAT affect the investigation? I don’t know but the timing seems peculiar that this is a big issue after the US divested itself of its GM stock.
There are a LOT of questions to be answered and I do not trust those asking the questions. Barra did a fine job under very difficult, politically charged questioning.
I guess that everyone hates Toyota, too, for stonewalling the unintended acceleration problem and its multiple causes, including the unsettled ones. And the were fined $1.4-billion to settle a criminal trial. Shouldn’t buy from them, either.
Thank you mustangman for the explanation from the inside. I fully understand the agony of anyone dieing in a car accident but I thought the lady that demanded that all of those cars be taken off the road was a little absurd. I still have the question of why the air bag did not works since the comment was made that the bag should work 60 seconds after the key is switched off.
Yeah, me too, @Bing. I have the same comment about the brakes. The brake booster must allow several power-off stops before the boost goes away. Unfortunately, I think this may be GM’s version of AUDI’s "unintended acceleration. I have no love for GM but they don’t deserve a witch-hunt. They are addressing the problem while Congress looks for the scapegoat.
To me, it’s utterly amazing that so much mayhem has been caused by ignition switches accidentally being turned off. You can still steer, you can still brake. Aside from the airbag not working x amount of time after power is interrupted, how is this any different than your car just stalling?
60s and 70s cars often had problems that resulted in stalling. (and no airbags and some without power steering and brakes) No one would have thought to sue because of this. I guess we’re all just pretty soft these days, not that GM shouldn’t fix this problem…
@mustangman, thank you for adding your inside perspective from Delphi. It definitely helps to have someone who understands the inner workings of GM explain how things might have happened.
But there are still unexplained things, like why the redesigned switch was not given a new part number. At the very least, it’s just plain terrible configuration management practice. Even Mary Barra was mystified:
“The document showed that the part number did not change, when redesigned, an issue which GM said hampered its own internal investigation.”
“It is inconceivable,” Barra said, when asked about the design change without a corresponding change in part number. “It is not our process.”
Also, it appears that the head switch engineer on the Cobalt, Ray Degiorgio, signed off on the switch design change in 2006, despite testifying under oath in a 2013 lawsuit against GM that he had no knowledge of the design change. So he may have some more explaining to do.
“The document has the signature of GM “lead engineer” Ray DeGiorgio, who could not be reached for comment. In a 2013 deposition in a suit against GM, DeGiorgio had said he was unaware of a change in the part.”
Quote from jesmed: “But there are still unexplained things, like why the redesigned switch was not given a new part number. At the very least, it’s just plain terrible configuration management practice.” Unquote
In response I can say that the company that I worked for also did not change part numbers when a part was modified. To identify modified parts, changes were assigned a letter such as PN 49-87349 per change A, B, C etc. and if Z was infrequently reached, then A1 would be next. Parts are always in a state of change and the paperwork needed to keep the system current with a constant flow of new part numbers would be formidable. Part numbers available are infinite but would get longer to the point of not being recallable from memory. People who work with part numbers can work more efficiently if they can memorize part numbers and for that to happen, part numbers need to be consistent and as short as possible.
One thing that is not often mentioned in the news or here, is the wearing of seat belts. It should be much higher on the list of considerations by all concerned. The young people from Wisconsin who were killed and injured were breaking the law by not wearing their seat belts.
Just for fun I turned the ignition of our 09 Cobalt completely off before I made the turn into our driveway. It surprised me how little additional effort was needed to turn the steering wheel a moderate amount but with sharper tuning the steering became much heavier but not uncontrollably so in my view.
The steering did not lock; requires that the key be pulled out of the ignition switch for that to happen.
The brakes worked normally but I applied them only twice, once before the turn and again to come to a complete stop.