Tech Search

When this April the driver's parents and a U.S. senator finally prevailed upon Toyota to examine the contents of the truck's crash data recorder, the electronic readings suggested a collision that was far from ordinary.

The data indicated the truck had been going 177 mph when it hit the tree, much faster than what the pickup possibly could go, safety experts said. Yet a separate reading from the recorder put the speed at 75 mph before impact.

As the National Highway Traffic and Safety Administration conducts its investigation into unintended acceleration in Toyotas, the crash data recorders or "black boxes" in the vehicles have become a primary piece of evidence. But long-standing reservations about the reliability of the data -- some of which have been voiced by none other than Toyota officials -- raise doubts about how much the safety agency can rely on them to determine the cause of the crashes.

Last week, NHTSA briefed Congress on its research and presented findings from 58 event data recorders, or EDRs. In 35 cases, the recorders showed the brake was never applied, suggesting that some of the alleged cases of sudden acceleration might simply have been driver error, agency officials said.

Department of Transportation spokeswoman Olivia Alair said NHTSA has conducted "validation tests to ensure that EDRs were providing accurate pre-crash data. As a result of that testing, NHTSA is confident that the EDR information obtained in those specific incidents is accurate."

But in the past, Toyota has expressed doubts about the recorders. Toyota declined to comment for this story.

In 2008, Toyota questioned the reliability of the devices in an effort to prevent the driver of a Toyota Echo from using data from the crash recorder in a lawsuit against the automaker.

"The data retrieved from the EDR is far from reliable," a Toyota court filing said at the time. "The EDR was not intended to be used a reconstruction tool in the field. It has not been validated as a reliable reconstruction tool or crash data recorder for crash events in the field."

Toyota's attorneys also noted that although both sides in the case agreed that one of the passengers in the vehicle was belted and the other was not, the data recorder said both seat belts were unbuckled at the time.

"Everyone agrees that's wrong," the Toyota attorney said, according to a transcript.

"The Toyota EDRs are so unreliable that even Toyota has challenged their reliability in court," said Clarence Ditlow, director of the Center for Auto Safety. "Given the demonstrated errors, NHTSA can't rely upon them in its investigation."