Skip to content
Link copied to clipboard

GM says recalled cars safe, but has not tested ignition for knee-bump danger

General Motors says that cars being recalled because of faulty ignition switches can be driven safely before repairs, based on more than 80 tests, but the automaker has not addressed a problem long known to potentially shut off the engine: a simple bump from a driver's knee.

WASHINGTON - General Motors says that cars being recalled because of faulty ignition switches can be driven safely before repairs, based on more than 80 tests, but the automaker has not addressed a problem long known to potentially shut off the engine: a simple bump from a driver's knee.

Safety advocates and engineers say the lack of testing for this factor undermines GM's claims that the cars are safe. As early as 2004, GM engineers complained that the ignition switch could be turned off if the key was bumped by a knee.

A Texas judge on Thursday allowed the unrepaired cars to stay on the road, over the objection of safety advocates and plaintiffs lawyers who said there is no way, short of repairs, to ensure the ignition switch would not slip out of the run position, turning off the motor and disabling power steering, power brakes and airbags. At least 13 people have died in such incidents.

GM, including CEO Mary Barra, has repeatedly said the 2.6 million cars it recalled for ignition switch problems are safe to drive, as long as they are driven with only one key on the key ring.

It is not clear if a driver's knee could still bump the ignition out of the "run" position when only a bare key is being used rather than a key ring with additional keys and possibly other things attached.

GM in court filings in the Texas case said it made more than 80 tests of driving with a bare key. It described tests driving over a pothole four feet wide by seven feet long by five inches deep at 25 miles per hour, driving up and over a 4-inch high simulated median at an angle and locking up the brakes while coming off the median, and driving a 4-mile loop "with a series of bumps, swells, railroad crossings" and other hazards at posted speeds of 25 to 75 miles per hour, for instance.

The filing described tests in which external forces bumped the car, rather than a jolt inside. There is no indication of a test for knee bumps.

Asked about the issue, GM told Reuters that none of the tests included a direct force on the key from inside the car, such as the driver's knee.

"Based on more than 80 individual tests, including some very severe tests like driving over a railroad crossing at high speed and driving over river rocks, potholes and cobblestones, we concluded that the recalled cars are safe to drive provided just the ignition key is used to operate the vehicle. The results of the tests, all of which are described in our affidavit, speak for themselves," GM spokesman Jim Cain said in a statement. The tests were run in March, he added.

Cain declined to say if GM would test for knee bumps in the future or further elaborate on the issue.

But knee-bumping complaints were among the earliest indicators of an ignition switch problem roughly a decade ago, and some safety advocates say GM should test for it.

"They should have run that test because that's one of the known failure problems," said Clarence Ditlow, executive director of the Center for Auto Safety, a watchdog group.

The switch is low on the steering column, which makes it more susceptible to bumps, GM drivers have said in company documents, and a spring in the switch which helps hold the ignition key in place does not push hard enough.

"They should be setting up some type of experiment that would have the key directly impacted by the knee," said Steven Batterman, a forensic engineer in New Jersey, who has worked for plaintiffs' lawyers in auto accident cases, including ones against GM.


In documents GM and parts maker Delphi Automotive supplied to congressional investigators and safety regulators, GM test drivers and customers over the past decade repeatedly experienced incidents when they accidentally bumped the key and turned off cars now subject to the recall.

In October 2004, Gary Altman, the program engineering manager for the two recalled cars, the Chevrolet Cobalt and the Saturn Ion, reported that he bumped the key with his knee and turned off the ignition while driving a 2005 model Cobalt, GM documents submitted to Congress and released publicly show.

"While driving the vehicle the drivers knee bumped the key in such a manner as to turn off the engine," one of the documents says.

In a GM document introduced last year in a Georgia lawsuit, 6-foot-3-inch GM driver Onassis Matthews said he inadvertently turned the ignition key off with his knee while test driving a Saturn Ion in February 2004. Matthews suggested moving the ignition key to a different location in the vehicle.

Meanwhile, GM in a February 2005 service bulletin warned its dealers to look out for short drivers, who would be more likely to bump the steering wheel column, according to a GM document made public in the Georgia lawsuit.

A second bulletin sent by GM to dealers in December 2005 show GM was concerned about drivers sitting too close to the steering wheel.

"In these cases, this condition was documented and the driver's knee would contact the key chain while the vehicle was turning and the steering column was adjusted all the way down," said the bulletin. In such cases, GM recommended removing extra material from the key chain.

It is unclear whether drivers or engineers reporting these early problems were using more than a single key.

The auto safety regulator, the National Highway Traffic Safety Administration (NHTSA), has advised drivers to follow GM's advice and drive with only one key. It did not respond to requests for comment about the risks of knee-knocking.


Safety advocates and engineers want more details of the tests that GM has performed. The court documents only describe the tests broadly.

Mark Hood, a forensic engineer in Florida who tested ignition switches on behalf of the Georgia family suing GM last year over the death of their daughter, said that is not enough.

He said it may be possible, under certain conditions, for a bump from the driver's knee to turn a single key in the weak ignition switches given the low level of force or "torque" it takes to do so. "I think it would depend on the angle of how they hit it," Hood said.

U.S. Senator Richard Blumenthal also has asked Barra to provide the data the company used to assert that the cars are safe. Blumenthal has advocated on the Senate floor for pulling the cars off the road until they are repaired.

"The plain simple truth is jarring or overloading the key ignition switch can shut down the car," Blumenthal said in an interview with Reuters.