Days after a driver was killed when his Tesla Model X crashed near Mountain View, Calif., a National Transportation Safety Board investigation has turned to the subsequent vehicle fire and a defective crash shield on the stretch of U.S. 101 where the wreck occurred, authorities said Wednesday.
While questions have been raised over whether the SUV was in “Autopilot” mode, a semiautonomous feature offered by Tesla that has garnered attention — especially in the wake of a fatal autonomous Uber crash in Arizona last week — NTSB investigators say the focus of their probe is determining why the SUV caught fire and whether the wreck would have unfolded differently had a crash “attenuator” been working as intended. The barrier serves as a buffer between vehicles and the concrete median.
The NTSB is investigating whether the barrier, which is meant to work as a type of shock absorber, could have played a role in the crash’s severity. Photos showed extensive damage to the SUV, which was charred and appeared to have had its front end ripped off.
The 38-year-old driver was pulled from the wreckage and died later at a hospital. Police said the vehicle was traveling at “freeway speed.”
“We have never seen this level of damage to a Model X in any other crash,” Tesla said in a blog post this week. Tesla attributed the crash’s severity to the highway component.
The auto manufacturer, whose stock plunged in the wake of the crash and investigations, said it was “deeply saddened” by the March 23 crash. The SUV was so severely damaged that the company has been unable to obtain the vehicle’s data logs, Tesla said in the post. The NTSB said Wednesday it had recovered two pieces of equipment — the restraint control module that contains information about seat belts and the deployment of air bags, and a central “Infotainment” module with other information. But it was not immediately clear what information could be learned from the components. The SUVs sell for about $70,000.
NTSB spokesman Chris O’Neil said Wednesday that the crash attenuator, overseen by the California Department of Transportation, had been damaged in a previous incident.
“It had not been reset,” O’Neil said, adding that officials were “looking at the damaged attenuator and looking at an undamaged one and looking at if it had an effect.”
A transportation department spokesman reached by phone Wednesday said the agency is looking into the matter.
Meanwhile, O’Neil said it was unclear whether the car was being manually operated or controlled with an assisted driving feature known as Autopilot. Unlike autonomous driving, Autopilot offers a more limited set of features, such as cruise control that adapts to the speed of cars ahead and automatic steering. Tesla has said drivers using the feature should remain attentive and keep their hands on the wheel, so as not to become over-reliant on the technology.
“Do we want to know whether it was in Autopilot mode or not? Yes, we want to learn that,” O’Neil said. “When we learn that, will we pursue other lines of inquiry? I think the answer to that would be, ‘Does that knowledge present a particular safety issue?’ Because the mechanics of the accident [aren’t] fully known.”
The NTSB said two investigators were on the scene. The federal panel, which methodically examines railroad and plane crashes and highway incidents that can turn up broader insights on transportation safety, said that although vehicle-related fires are not uncommon, the relative newness of electric car technology made the crash a focus. Investigators will look into how the fire contributed to the difficulty of emergency crews’ crash response, and whether components intended to control the flames worked as intended.
“Here we have an electric vehicle involved in a post-crash fire — let’s explore ‘How did that happen?’ ” O’Neil said. “Did the batteries play a role in that? Did the batteries make it harder for the fire to be put out? Did that make it worse, harder to be transported?”
Tesla has said that its batteries are designed with firewalls intended to limit how quickly flames can spread.
“Serious crashes like this can result in fire regardless of the type of car, and Tesla’s billions of miles of actual driving data shows that a gas car in the United States is five times more likely to experience a fire than a Tesla vehicle,” the company said in its blog post.
Tesla owners have turned on Autopilot on the same stretch of highway about 85,000 times since the feature launched in 2015, the company said, and drivers navigate the road on Autopilot about 200 times per day. In that time, said Tesla, “there has never been an accident that we know of.”
The Associated Press contributed to this report.