July 2, 2010
We sat through the National Academies of Science first public meeting to tackle the Electronic Vehicle Controls and Unintended Acceleration Study, a NHTSA-sponsored effort to look broadly at the issue, and we are happy to see that the agency has brought in some outside expertise.
This is truly an opportunity for the regulators to advance their knowledge base beyond the era of the mechanical automobile and into the age of automotive electronics, rapidly migrating from a vehicle’s entertainment center to its most basic functions of acceleration, braking and steering. It is critical to future policy setting and defect analysis.
The first day featured a host of presentations from NHTSA, and some of the good news is: the agency is focusing on the 2002-2006 Camry, which has long been known to be one of Toyota’s most troubled vehicles in terms of consumer complaints. NHTSA’s Roger Saul and Mike Kirsch of the National Aeronautics and Space Administration, which is partnering with the agency, presented a PowerPoint that showed the huge spike in unintended acceleration complaints beginning in 2002, when Toyota introduced its Electronic Throttle Control System-intelligent (ETCS-i). (SRS and Quality Control Corp. actually mined these data many months ago and came to the same conclusion. For that matter, the agency essentially knew this in 2003, when Steven Chan wrote an IE memo noting this same spike, albeit in a shorter time frame.).
Kirsch, a NASA systems engineer, laid out the scientific approach the team expected to bring to the process of ferreting out weaknesses in Toyota’s electronics, including studying complaint symptoms, understanding how the design is supposed to work, how it might fail, developing event sequences and fault trees, and finally, testing.
Among the automotive electronics experts who made presentations to the panel were Michael Pecht, a mechanical engineering professor who heads the University of Maryland’s Center for Advanced Life Cycle Engineering (CALCE) and who advised Henry Waxman’s committee on Toyota SUA, and Todd Hubing, an EMC expert andprofessor of vehicle electronics at Clemson University.
Pecht was openly critical of NHTSA’s past work on unintended acceleration as well as the industry’s inability to identify root causes for defects where no faults were detected by the vehicle diagnostics and their inadequate testing standards. Pecht also described his experience helping uncover intermittent electronic defects for the industry – including a major automaker whose cruise control modules were causing SUA – in which 96 percent of the returned modules tested fine despite huge numbers of field returns. Standards for testing are 40 years old and they have gotten more relaxed and the testing is not being evaluated in situ, according to Pecht.
Pecht also pointed out the need for NHTSA to find better ways to obtain relevant documents from automakers to advance their investigations and that the agency “bible” on SUA, referred to as the “Silver Book,” was lacking and biased toward human error.
Also presenting on the second day was Robert Strassburger from the Alliance of Automobile Manufacturers who gave an entirely general overview about the robust testing by the industry but was scant on details and instead referred the panel to the individual members of the association.
Hubing described his testing and failure analysis work for the industry and attorneys. Hubing was cautious in his assessment of the vehicle electronic issues that can result in SUA, taking particular pains to note that the odds of an SUA event is lower than the odds of being involved in other types of crashes, and SUA events are rare occurrences in general. But, he did note that the problem will get worse without a major change in standards and design practices.
One point that caught our attention was Hubing was able to replicate Dr. David Gilbert’s work and obtain wide-open throttle without the fault detection system setting an error code – but with only a single fault. Gilbert’s analyses found that first a loss of signal redundancy at the accelerator pedal sensor was needed followed by a voltage spike to create an unintended wide-open throttle. Hubing went on to note that many of the faults created invalid signals that sometimes would be detected, other times not. His findings were previously provided to NHTSA and NASA investigators.
But we are also a little anxious about how the panel will complete such an absurdly broad charge in a little over a year. NHTSA has asked the panel to cover the complete gamut of SUA causes. In addition, the assembled panel includes a number members retired from their careers. This is a regular feature of such National Research Council panels, because all are volunteers who must have sufficient availability to complete the task. For this reason, many of the SUA panelists are not necessarily on the technical cutting edge of automotive electronics. Also, it was clear from many questions posed by the panelists that they have a steep learning curve in understanding NHTSA’s processes or even in understanding the history of this defect.
It did not help that NHTSA Administrator David Strickland told the panel in his opening remarks that the agency had already identified two “known” causes of unintended acceleration in Toyotas – floor mats and sticky accelerator pedals. Even Toyota – which has enjoyed some of the cover that the false linkage between its sticking accelerator pedal recall and unintended acceleration has provided – has acknowledged to Congress that it is not a cause of unintended acceleration in three-quarters of the cases out there.
We are also dismayed by the amount of attention focused on human factors. This is already a well-worn path for the agency. NHTSA actually doesn’t need any outside help on human factors and unintended acceleration – it wrote the (Silver) book on it (The 1989 electronics-denier’s Bible: An Examination of Sudden Acceleration). As presented Wednesday, the human factors influence on unintended acceleration sounds very reasonable, until you actually start reading the Toyota complaints and talking to owners who’ve had multiple instances in the same vehicle with different drivers, or witnesses who observe a crashed Toyota continuing to rev its engine with no pedal application, or high-speed events where the driver is already applying the accelerator.
It would be disappointing in the extreme if the agency wasted this crisis by retreating to its comfort zone, and concluding: See, we told ya it’s those dumb drivers!
In Other Washington News…
Rep. Henry Waxman’s Energy and Commerce Committee is very busy these days with British Petroleum’s destruction of the Gulf of Mexico, but they haven’t forgotten Toyota.
This week, the Commerce Committee’s Sub-Committee on Oversight and Investigations sent Toyota Motor Sales CEO Jim Lentz a letter apprising him Exponent’s non-cooperation in supplying the committee with key documents. In particular, the committee was miffed that Exponent, by admission of its own representative Dr. Shukri Souri, “had withheld from the Committee a ‘living document’ which he stated was ‘an incomplete set of our opinions and conclusions’ that is ‘continuously updated that is reflecting our progress.’”
Instead, Exponent submitted a substantially altered document. The Committee reminded Toyota that, despite the baffling the automaker wrapped around Exponent by hiring it through litigators Bowman and Brooke, the Committee still regarded Toyota as Exponent’s minder and asked the automaker to direct its contractor to Giddy-up.
More interesting was a request for information about a brake override mechanism that unidentified technical experts found on a 2005 Toyota Camry.
Astute hearing watchers may recall that Lentz told the Committee under oath at a February hearing that the company could not install a brake override function on the older model Toyotas, because it wasn’t technically feasible. Here’s an exchange between Sub-Committee Chair Bart Stupak and Lentz:
REP STUPAK: Well, on this rebooting of this computer there, why only some of the vehicles are going to be rebooted and not the other ones?
MR. LENTZ: It depends on the feasibility of the unit. Some of the computers are different types of chips in them.
REP. STUPAK: Okay.
MR. LENTZ: Some are not rewritable, basically. They’re hard coded.
REP. STUPAK: So can’t you fix those — rewrite the program so all your vehicles are covered? I mean, what do you say to these owners who are not going to have this safety feature added to it? They’re just going to continue to drive down the road and hope they don’t have a sudden, unintended acceleration?
MR. LENTZ: Well, again, the incident or the possibility of that happening is very, very slim. But understanding if it happens to you, it’s a very, very important incident.
Also very, very important: Not BS-ing the Committee.
Apparently, some older model Camrys, not included in brake override recall, already have some form of override function.
According to the letter:
“We have been notified by outside technical experts that based on their examination of a 2005 Toyota Camry, some older model Toyotas may have a brake override function that activates when the vehicle’s electronic throttle control system registers a diagnostic trouble code. These experts report that this brake override function is separate from ‘limp home’ mode. In your testimony before our Committee – on February 23, 2010, and again on May 20, 2010, you described Toyota’s commitment to outfit all vehicles with a brake override function beginning with model year 2011 and to retrofit some older model vehicles that are currently on the road. On April 28, 2010, electronics experts from Toyota briefed Committee staff on brake override features in Toyota vehicles. Neither you nor Toyota’s electronics experts informed the Committee that some older model vehicles may already have a brake override function that is tied to a diagnostic trouble code.”
Question: Did NHTSA know about Toyota older brake override system? Or is this another one of Toyota’s little secrets?