Subscribe to Print and Digital for only $109!

Critical question went unasked by NHTSA, unanswered by GM

How is ignition position linked to airbag deployment?

WASHINGTON -- At this week’s congressional hearings, watch for General Motors CEO Mary Barra to apologize deeply for her company’s handling of deadly defects in its ignition switches. And watch for David Friedman, the nation’s top auto-safety regulator, to make no apologies at all.

Friedman, acting administrator of the National Highway Traffic Safety Administration, is expected to testify that the peculiarities of emerging “smart” airbag technology made it difficult for NHTSA to figure out whether something was wrong with the Chevrolet Cobalt and Saturn Ion during a 2007 inquiry, and that it was GM’s fault for withholding critical clues to that puzzle.

“This was a difficult case … involving airbags that are designed to deploy in some cases, but not in cases where they are not needed or would cause greater harm than good,” Friedman says in prepared remarks released ahead of a U.S. House Energy and Commerce Committee hearing today. “GM had critical information that would have helped identify this defect.”

Yet even now, the agency cannot say exactly why GM's ignition switch defect caused these airbags not to inflate. And it seems that it was not until after the February recall of some 1.6 million cars, and at least 13 deaths linked to the faulty switch, that NHTSA asked GM about it.

Assumptions about power

GM has now recalled 2.6 million cars globally to fix a flimsy ignition switch that could easily be jarred out of the “run” position and into “accessory” mode. GM says that could not only cut off the engine power, but also deactivate the airbags.

But a lack of power isn’t the problem. Airbag systems are commonly built so they aren’t entirely reliant on engine or battery power, with capacitors that store a few seconds’ worth of backup power to deploy the airbags even if a car’s engine and battery are obliterated in a crash.

The agency’s “understanding of airbag systems, which was verified by available GM service literature reviewed during our due diligence effort, was that an airbag system would be armed and ready to fire for up to 60 seconds after all power to the system was cut off,” Friedman says in his prepared testimony.

Emergency crews know this. This is why when they arrive at a crash scene, they turn off the ignition, cut power from the battery and wait a couple minutes for the capacitors to die, so that no one is injured by a late-firing airbag, says Ron Moore, a first-response trainer from Texas and a contributing editor of Firehouse, a website frequented by firefighters.

In the case of the accidents linked to GM’s recall, there’s little question that the airbags had enough power available to deploy. The engine went off, but just seconds before the crash. In the fatal Cobalt crashes for which NHTSA conducted "special crash investigations" to understand why the airbags didn’t deploy, the battery had enough juice to power a “black box” in the last few seconds before impact.

Even if the battery had failed, the capacitors should have saved the day.

So why didn’t the airbags fire?

'Is everything still going to be alive?'

Here is the nuance that NHTSA seems not to have explored until the recall erupted in February: While lack of power is not generally the reason that airbags fail, there are other factors -- and the position of the ignition switch can be a big one.

Experts say automakers have varying policies that dictate how their safety systems are programmed to behave, depending on the position of the key and whether the engine is turned on or off.

Joe Nolan, chief administrative officer at the Insurance Institute for Highway Safety, says the group asks automakers about those policies every time it conducts a crash test at its laboratory in rural Virginia. Crashing a car is a major expense, and the IIHS needs to make sure vital safety systems like airbags will all be activated, he says.

“We ask them: If we put [the car] in ‘run’ and have a 15-to-20 minute delay between doing that and crashing the vehicle, is everything still going to be alive?” Nolan says. “Sometimes, based on what they tell us, we have to change the protocol.”

Automakers treat the “accessory” position in different ways, Nolan says, because the purpose of the accessory position is to power some of a car’s systems -- the radio and the windshield wipers, for instance -- without draining the battery.

Early warning

Nolan was recently sitting in a parking lot in a Volvo XC90, trying to listen to a broadcast of his alma mater, the University of Virginia, in the NCAA men’s basketball tournament. The key was in the accessory position. He had the radio on, and the engine off.

Within half an hour, Nolan says, a low-battery warning popped up on the dashboard.

Nolan was surprised. Half an hour seemed too soon for a battery warning. When he put the question to Volvo, though, the company told him it was by design. Volvo vehicles keep running safety systems like airbags in case of an emergency, even at the expense of battery life.

Nolan says it’s conceivable that GM programmed its airbags not to deploy with the key in the accessory position. Perhaps the company wanted to save power, or cut down on false deployments.

“I can’t say specifically how GM is doing it,” he says, “but different automakers seem to be doing things in different ways.”

NHTSA’s questions

Friedman will testify to Congress this week that NHTSA dug into the data on Cobalt crashes in 2007, when it was receiving investigation reports on fatal accidents in which the airbags didn’t deploy and the ignition was in the accessory position.

The agency found 43 Cobalt crashes in which airbags might not have deployed, his testimony says, but the car didn't seem to be an outlier. Compared with the 2005-06 Cobalt, some models had lower rates of injuries in accidents in which airbags didn’t deploy. But others -- including the 2004-05 Kia Spectra and GM's own 2004-05 Chevy Aveo -- had higher injury rates.

"The available data did not indicate that the Cobalt or Ion were over-represented compared to other peer vehicles," Friedman will testify.

NHTSA doesn’t seem to have asked GM to explain whether nudging the key into the “accessory” position could deactivate the airbags -- a question raised by a special crash investigator in an April 2007 report on a fatal crash in Wisconsin.

In a chronology filed with NHTSA last month, GM said it conducted in-house testing in 2013 on that point. The testing revealed that airbag deployment depends on a number of factors, including “the timing of the movement of the key out of the 'run' position relative to the activation of the airbag's sensing algorithm of the crash event."

NHTSA wasn't satisfied with that explanation. In a 107-question query on March 4, the agency asked GM to explain more about its airbag system.

The query asks GM to explain "whether or not GM intended for the subject vehicle frontal airbags to deploy in a crash when the ignition switch is in the accessory position, or in the off position, or in an interim position, and describe any additional conditions or factors that may affect whether or not the [sensing and diagnostic module] disables the frontal airbags.”

That’s the question NHTSA apparently didn’t ask in 2007. An answer is due Thursday.

You can reach Gabe Nelson at

ATTENTION COMMENTERS: Automotive News has monitored a significant increase in the number of personal attacks and abusive comments on our site. We encourage our readers to voice their opinions and argue their points. We expect disagreement. We do not expect our readers to turn on each other. We will be aggressively deleting all comments that personally attack another poster, or an article author, even if the comment is otherwise a well-argued observation. If we see repeated behavior, we will ban the commenter. Please help us maintain a civil level of discourse.

Email Newsletters
  • General newsletters
  • (Weekdays)
  • (Mondays)
  • (As needed)
  • Video newscasts
  • (Weekdays)
  • (Weekdays)
  • (Saturdays)
  • Special interest newsletters
  • (Thursdays)
  • (Tuesdays)
  • (Monthly)
  • (Monthly)
  • (Wednesdays)
  • (Bimonthly)
  • Special reports
  • (As needed)
  • (As needed)
  • Communication preferences
  • You can unsubscribe at any time through links in these emails. For more information, see our Privacy Policy.