I’m listening to the hearings now and they are discussing the capacitor firing the air bags when the ignition is off. So there still is a question of why the bag didn’t go off and if that is a separate issure or related to the ignition. Sure looks to me like they are grand standing for the victims families though not fact finding. Mary was treated quite poorly in my view.
One morning I lost my serpentine belt due to a water pump. I managed to man handle the car off the freeway, turn around and back about a mile to a shop, all without power steering. Wasn’t easy and my arms were sore but still didn’t hit anything.
@wha who, different companies manage design revisions differently. Yes, there’s a matter of judgement as to when a new part number is issued. Commonly a minor revision can be identified with a change in dash number, so there doesn’t have to be a total part number change, only a new dash number at the end of the part number. But there has to be some change to the part/dash/revision number marked ON THE PART so that it can be distinguished from the previous dash number or revision number.
That did not happen here. There was absolutely no new dash number or revision number marked on the parts with the changed design. GM themselves can literally not tell the difference between an “old” switch and a “new” switch by looking at it because they are marked identically with the same part/dash/revision number.
Mary Barra herself said that it was “inconceivable.” So there’s no excuse. Someone either totally screwed up or was actively trying to hide the fact that the design had been changed.
It’s been my experience that when a part is superseded it’s assigned a different number and that may vary from a single digit or letter to something completely different.
What Barra had to say is boilerplate written by the legal staff. Those members of Congress know what someone is going to say before they even hit the city limits sign. It’s all a dog and pony show.
Barra said they will find out whodunnit. Gambling money says nothing will happen outside of the Recalls other than some families being paid.
Gambling money also says a lot of those GM execs who cashed in their stock options and bailed after the Feds took over were in on this also.
My gambling money says the guy who signed off on the switch redesign, without changing the part/dash/revision number, and then apparently lied under oath by saying he had no knowledge of the redesign, will be one of the fall guys, deservedly or not.
I still think they should somehow make Bob Lutz the fall guy
He’s such a prominent and arrogant jerk, the public might be somewhat satisfied if he was the loser at this game of musical chairs that is now being played
Adding a dash number to a part number such as 49-3374 to make it 49-3374-1 makes it a NEW part number. That means that if there are, for example, 50 assembly drawings that specify that part, then all need to be changed to show the new part number. That is not a way to minimize paperwork. Each changed drawing will need to go through a company’s internal release system for approvals from various people who will check the accuracy of the change but may not know why the dash number was added. Identifying a part with a letter designation such as 49-3374 per change A before and per change B after a change makes it so only one drawing needs to go through the system.
Then a decision needs to be made; use, repair or scrap the old parts. It will be interesting to know how GM handled this. Not using a new part number for a lightly modified part is not intended for hiding anything.
You’re missing the whole point and getting lost in the semantics of what constitutes a “new” part number. Which, as a retired mechanical engineer, I do understand.
The point is that GM literally did NOTHING at all to distinguish between the old and new switches. The “old” parts and the “new” parts have exactly the same designation. Not a dash number change, not a “Rev A” or “Rev B”. Nothing. And now they can’t tell the old switches from the new switches. It may or may not have been intended to hide something, but at the very least it was incompetent. And if it wasn’t intended to hide something, why did the engineer who signed off on the design change lie about it under oath?
Mary Barra herself said that the way the redesigned switch was not differentiated from the old switch in terms of identification was “inconceivable.” That’s from a lifelong GM engineer who presumably knows how their system is supposed to work.
Why is this a 10-year-old problem? Our first (of several) GM cars with the key in the side of the steering column was a 1979-1/2 Pontiac Phoenix, a very early X-Car. I used an extremely heavy key ring at that time, a practice I have since discontinued. One day in 1984 my wife called from work when the car wouldn’t start. I managed to get it started with the aid of a large screwdriver as a conductor to power the starter. We took it to the dealer, paid to have the key actuator replaced, and I stopped using the massive keyring. The actual switch, by the way, is farther down the steering column. We had no problem with subsequent GM cars (1982 Pontiac Phoenix, 1983 Buick Century, 1984 Oldsmobile Ciera, and 1984 Pontiac Firebird). The key actuator from the first Pontiac Phoenix looked pretty much like the current questionable one I just saw on TV. Nothing REALLY bad ever happened. So how is this defective ignition switch design a problem with GM cars just in the last decade when I first encountered it in a 1979-1/2 Pontiac?
It’s not really a big problem if your wife’s ignition switch won’t work in the parking lot.
It’s a bigger problem if her ignition switch accidentally turns off at 50 mph, she loses control and crashes and is killed, which is what has happened to many people in this case.
The difference is that GM/Delphi made the switch with very little resistance force to turning off, which made it very susceptible to accidental shutoff by jostling it or carrying too much junk on the keychain. Their fix is to make the switch harder to turn off…like all other ignition switches that work well and don’t turn off by accident.
Why they made this particular switch so easy to turn off in the first place is something of a mystery. It was so easy to turn off that it didn’t even meet their own standards, but they went ahead with production anyway.
The news reports stated that there was a revised detent spring and plunger assembly that was about 15% longer than the original. That extra length of course would mean the switch would have remained more firmly in place.
This would be similar to a manual transmission in which a shift rail detent spring has failed or someone substituted the wrong spring during assembly work. The transmission may jump out of gear for the same reason that the ignition switch may flop out of position.
The upshot to all of this is that no one will go to jail, fines will not likely be assessed except as a token gesture, the families will be paid to go away, no one in the NHTSA will be held accountable, and nothing will change except Ms. Barra having a short tenure.
At some point this kind of thing will resurface again and involve another component besides the switch.
Some commentators have speculated that the board chose Mary Barra because they knew this crisis was coming down the pike, and women are better than men at cleaning up messes and apologizing to victims families than men are…and that as a result she would have a short tenure…the so-called glass cliff.
I truly hope that she wasn’t set up to fail, because I respect her credentials and think she could possibly turn things around at GM. But I’ve learned never to underestimate how cynical corporate boards can be.
jesmed, I agree that if GM used no method whatsoever to identify old vs new parts, that is unacceptable. It’s possible that the detent plunger and spring part numbers had revision letters to identify old vs new but I have not seen or heard that mentioned in the news. The same part numbers would, however, be used in the switch assembly. Assembly drawings seldom identify individual part drawing revisions. An assembly is also a part but it’s drawing would not show a revision unless one of the assembly’s part’s numbers was changed or if a note on the drawing was added, deleted or changed. That all is not an intentional coverup but in this case it may appear that way to some unfamiliar with engineering drawing procedures.
A decision on what to do if GM and the NHTSA recognized a potential problem would involve a huge recall which we are seeing now. Against issuing a recall, it could be reasoned that most people don’t hang an overly large set of keys and other things from their ignition switch, the car can be steered and easily stopped with no engine power if one does not pump the brakes but with more difficulty if they do, most people will have their seat belts fastened as is required by law or common sense in case something bad does happen, and with some presence of mind a few might even turn the ignition back to the ON position.
Figure the odds on all of that working or not! With the cost pressure of a huge recall and everything being a compromise between cost and perfection in the engineering world, my guess is that you would have concluded the same as GM engineers and NHTSA.
" With the cost pressure of a huge recall and everything being a compromise between cost and perfection in the engineering world, my guess is that you would have concluded the same as GM engineers and NHTSA."
As a retired engineer, and someone who cares about making quality products that don’t kill people, I hope I would have recommended redesigning the switch and recalling all cars using it as soon as I found out that (a) it didn’t meet the torque spec and (b) fatal crashes were occurring in which the switch was found in the “accessory” position and airbags were not deploying.
But I would have been overruled by the bean counters who didn’t want to add 90 cents to the cost of the car and would rather pay to settle lawsuits from the families of dead people.
Of course just my own lowly opinion and others will disagree, but I think the way the senators have treated a major corporate CEO who is newly on the job and attempting to drill down to what happened, is disgraceful. Among those of the worst are Boxer, Rubio and the senator from NH, MA, Mo, and others. I was heartened to see our own Senator Klobuchar from Minnesota keep her questions short, respectful, and relevant. The political grandstanding is so transparent, they should just watch themselves on TV once to see how they really come across. “We need to have a law so that this will never ever happen again”. Yeah sure, you folks have done a fine job so far. It sure must be GM’s fault that rental car companies don’t send their cars in for recalls, yada, yada, yada. And have you fired this man yet?
Maybe the old boys at GM just want her to take the fall so they can put Lutz’s cousin in charge or something and the ole congress just makes it easier. How about instead a little support into trying to find out what actually happened and why and a little soul searching instead on both sides?
Also, I’d like to see some pictures of massacred cars in which the belted, sober, driver lived because of the features of the car. As i write my check again to the IRS, it bothers me a little that I am paying for these fools.
You can always count on Congress to put on a good show of beating up whoever appears before them to score points with the folks back home…whether or not that person had anything to do with the scandal du jour.
Speaking of the blame game, here’s a good point about Congress’ own failure:
With NHTSA’s budget for defect investigation around $11 million per year, and around 16 million cars sold per year in the US, that’s less than 70 cents per vehicle devoted to defect investigation. 70 cents won’t even buy a Snickers bar.
The congress question session is just a show. I don’t watch it, if I wanted to watch, Judge Judy might be more relevant.
Corporations are there to make money, they have bean counters. Sometimes they make stupid decisions. I see this very often. The sad part is, the ones who end up paying, are not the ones who profited from the greed.
Has anyone dissected one of these failed ignition switches? I am interested in the mode of failure i.e. where the engineering failed. Were the contacts not robust enough? Did resistive heating change the contact pressures? Is there a TSB about this situation?
It is amazing that a forensic engineer has not come up with a smoking gun for these failures.
@researcher, yes. The first “outsider” to figure it out was an engineer hired by one of the victims families. He dissected the ignition switch from the car crash and compared the innards to later switches. He found that the design had been changed at some point after his client’s car had been made.
The problem was that the earlier switches did not have enough resistance to accidentally turning the switch off by jostling it or carrying too much junk on key rings. The switches just accidentally got turned to the “accessory” position while the car was being driven. In that position, the engine shuts off and the airbags become inoperable.
The accidental engine shutoff while driving was so common that many drivers experienced it in non-fatal conditions and lived to complain about. Only a small percentage of the accidental shutoff incidents led to a fatal crash. And some of those families sued GM. Invariably the ignition switch in the crashed car was found in the “accessory” position and the airbag had not deployed.
GM redesigned the switch in 2006 to provide more resistance against accidental rotation. But the redesign was done so quietly that many people in the company who should have been involved didn’t know about it.
So there’s no “mystery” now. Everyone knows that the old switches were just too easy to turn off by accident.
GM knew from the beginning that the switch did not meets its own torque specs. The supplier, Delphi, told them so. But for some unknown reason, probably cost, GM decided to go ahead with production anyway.
