CARBONDALE, Ill. • None of this happens if David Gilbert keeps his Ford F-150 truck.
Not the threats to his job as a professor at Southern Illinois University Carbondale. Not the corporate intimidation. He never would have testified before Congress. And he never would have met those NASA scientists, the ones who ended up naming an electrical effect after him. As a gift, they gave him the black NASA coffee mug sitting on his desk — the one he’s sipping from right now. The mug reads, “If it’s not safe, say so.”
And that’s what he did. Four years ago, reports of crashes caused by the sudden unintended acceleration of Toyotas filled the news, just as stories about faulty ignition switches in GM vehicles do now. Back then, the Japanese carmaker recalled millions of vehicles. And Gilbert, who teaches automotive tech, got curious. He had just traded in his F-150 for a Toyota Tundra. So he pulled his new Toyota into a class shop and went hunting for a culprit.
What he found — that complex electrical controls could be to blame — remains controversial even today. His findings are viewed as either bad science or an unresolved smoking gun, the danger still lurking on the nation’s roads. Last month, the U.S. Justice Department settled criminal charges against Toyota for $1.2 billion, a huge fine but one that dealt with only two causes of runaway vehicles: bad floor mats and sticky pedals. Yet Gilbert’s basic theory continues to have supporters, and it is being tested in lawsuits across the country.
It would have been easier for Gilbert to stay silent. He nearly did. He hesitated before sending his report to federal regulators four years ago, reading his letter again and again, his finger hovering over the mouse, waiting to push “send.”
“I knew when that email went,” he says now, “that it was going to be different for me.”
He had no clue.
Gilbert, 55, grew up on a farm in Oklahoma. He has always loved vehicles, first tractors, then cars. He went to college to learn to show others how to work on them. At Southern Illinois University, he teaches a course on electronic engine controls.
Modern cars are filled with computers. Even things that look manual, such as the gas pedal, are really just platforms for electronics to relay messages. In newer Toyotas, and many other vehicles, pressing the gas pedal triggers a sensor, which signals another computer how wide to open the engine’s throttle.
The problem with Toyotas popped into public view with the high-profile 2009 crash of a Lexus ES 350 in San Diego. (Lexus is Toyota’s luxury brand.) Four people died when the car sped out of control. Investigators blamed an accelerator pedal trapped under a floor mat.
Other cases followed. Gilbert heard about them on the TV news. But he wondered if the problem might also be electrical. He hooked up his Toyota Tundra to a machine he uses in class. He was able to trick the vehicle’s computer into throwing open the engine throttle by inducing an electrical fault. It was a simple problem and one that should have been detected by the vehicle’s computer, triggering an error code. But the fault was invisible to the vehicle’s computer brain. To the computer, it appeared that a driver was stomping on the gas pedal, even though it was just two shorted circuits.
“I couldn’t explain how it happens,” Gilbert recalled, “but I could do it easily and not set an error code.”
He wrote up his findings and sent them to the National Highway Traffic Safety Administration and Toyota. He figured he was done. Then he called one of the people he saw quoted in a TV report on the runaway cars: Sean Kane with Safety Research and Strategies Inc., in Massachusetts.
Kane was stunned. Experts across the world had just assumed a vehicle’s computer would detect a shorted pedal sensor.
“It opened the door for everybody,” Kane recalled.
In February 2010, Gilbert got a phone call. Did he have a black suit? Could he fly to Washington to testify before Congress? He was on a plane the next day. He had to run out and buy his first cellphone.
Gilbert faced a flood of cameras as he sat at a long table in front a House oversight committee. He didn’t know it then, but Toyota was developing a public relations campaign to attack his credibility. Gilbert was nervous as he read from his prepared statement. Then Congress members began peppering him with questions.
He started to relax. He felt like he was leading a class.
“I went into teach-mode,” he recalled.
Gilbert returned home as a celebrity. Before this, he wasn’t even the best-known “David Gilbert” on the 20,000-student Carbondale campus. That title belonged to a psychologist.
Now, Gilbert was receiving phone calls from around the world. People showed up at his house to talk about cars. Toyota officials visited campus for a meeting with Gilbert and his bosses to go over his congressional testimony line by line. “They were steaming,” Gilbert recalled.
The carmaker hired experts to show that Gilbert’s results would never occur under real-world conditions. The problem was mechanical, not electrical, the carmaker insisted.
It was a delicate matter. Toyota was a big supporter of the school’s auto-tech program. Two Toyota workers resigned from the program’s advisory board. A vehicle donated by Toyota for classwork was taken back. In both cases, the carmaker said the moves had nothing to do with Gilbert’s testimony.
The administration tried to block him from working further on the acceleration issue, until he filed a grievance.
“If I didn’t have tenure, I would’ve been on a lot shakier ground,” he said.
Kane, the safety consultant, was more blunt.
“It’s disgraceful how they treated him,” he said. “He should’ve been lauded as a hero.”
But Gilbert was hailed by scientists at NASA’s Goddard Space Flight Center in Maryland. They were captivated by his testimony about an electrical problem with seemingly no obvious cause.
These scientists studied tin whiskers.
Tin whiskers are easy to miss, thinner than a human hair. They look like metal fuzz. They grow — for reasons scientists don’t understand — from plated tin surfaces, millimeter by millimeter. And if they bridge two closely spaced circuits, tin whiskers can cause a short.
Tin whiskers can wreak havoc in electronics ranging from simple digital watches to modern space satellites, from pacemakers to nuclear power plants. NASA studies them closely. And the problem in Toyota’s pedal sensor sounded like it might be caused by tin whiskers, said NASA physicist Henning Leidecker.
Leidecker and other NASA scientists were so taken by Gilbert’s research that they call the unique sequence of events required for a pedal sensor to short out “the Gilbert Mechanism.”
“I think he’s a hero. What he found was ingenious,” Leidecker said.
Federal auto regulators decided to investigate whether tin whiskers played a role in Toyota’s acceleration problems and asked for NASA’s help. In 2012, the agencies issued a report finding no electrical cause. Tin whiskers were ruled out. Toyota called it “a discredited theory.”
But Leidecker, one of the investigators, said the findings were more limited. Tin whiskers are elusive and easily destroyed. A case of a Toyota with galloping acceleration — but not a fully open throttle — was attributed to tin whiskers. And Leidecker pointed out that Toyota redesigned its pedal sensor in 2007 and again in 2008, expressly to eliminate the risk of tin whiskers. Why would it do that if tin whiskers were never a problem? he asked. Toyota did not respond to multiple requests for comment.
Leidecker said he believes the tin whisker risk remains for Toyotas in model years 2002-2006. While the risk is small, it increases with time. “It’s a game of Russian roulette,” he said.
Two years ago, U.S. Sen. Charles Grassley, R-Iowa, cautioned that the federal investigation made a mistake in ruling out electronics as the cause of Toyota’s unintended acceleration problem. Grassley said whistle-blower information indicated that the investigation may have been too narrow. But Grassley’s push languished. “It’s frustrating. Was there something we’re missing?” an official in Grassley’s office said on the condition he was not identified.
The support of NASA has helped Gilbert. He keeps the coffee mug on his desk and, on a whiteboard, a framed photo of him at NASA visiting with the tin whisker scientists. He now teaches his students about tin whiskers.
While the controversy has mostly blown over, some people have not forgotten. Just last week, a comment on the school’s auto-tech program Facebook page questioned why Gilbert still had a job.
Omar Trinidad deleted the comment. He’s an assistant professor. He helped Gilbert with some of his early tests on Toyota vehicles.
“For a while we were thinking, are we cuckoo? Because everyone seemed to think we were foolish,” Trinidad recalled. “But when even NASA said we found something, it kind of gave us peace.”