SUA—the Problem That Can’t Be Waved Away

For the metallurgist in all of us. The problem of Sudden Unintended Acceleration in Toyotas and other automobiles is raising its ugly head again, even after NHTSA had declared that there was absolutely no electrical cause for it. The latest explanation is also one of the earliest raised—tin whiskers, metal dendrites that can form on electroplated tin:

Much of the debate over tin whiskers died down in February 2011, after Secretary of Transportation Ray LaHood called on teams of NHTSA and NASA engineers to search for the cause of Toyota’s problems. “The verdict is in,” LaHood declared emphatically after the scientists reached their conclusion. “There is no electronic-based cause for unintended high-speed acceleration in Toyotas. Period.”

After that conclusion was announced, the National Academy of Sciences said that all available data indicated there was no electronic or software problem on the Toyotas.

Still, the buzz about tin whiskers continued in some quarters, especially after a separate team of NASA scientists published a study (PDF) of Toyota accelerator pedals on the Internet in September 2011. The scientists said they found tin whiskers in an accelerator assembly, and “a tin whisker induced short was responsible for the failure of a 2003 Toyota Camry Accelerator Pedal Position (APP) Sensor based on a Dual Potentiometer Design.”

In his letter, Grassley cited that study and its author, Henning Leidecker of NASA Goddard, to support his claim that more questions needed to be answered.

In response to the latest volley, Toyota officials argued that tin whiskers may cause failure of pedal sensors, but they’ve never caused unintended acceleration. “We know that tin whiskers can form — we’ve never said they couldn’t,” Brian Lyons, a spokesman for Toyota, told us. “But when pedals fail — for whatever reason — the vehicle enters into a fail-safe situation, a limp-home mode.”

While I would be completely unsurprised if most of the incidents of sudden unintended acceleration were, indeed, caused by driver error, I am 99.9% certain that some of the incidents can’t be explained that way. Or by floor carpets or mechanical linkage problems.

My own preferred explanation is software problem. The notion that such a problem could unquestionably be identified by code inspection alone (which is what Toyota and NHTSA have done and said) is laughable.

16 comments… add one
  • PD Shaw Link

    I don’t believe there is a metalurgist in me, I’ve looked. But if your talking about a mysterious 0.1% unexplained phenemena in manufacturing then your talking about gremlins and you’ll never be able to find them or explain them if your unwilling to recognize them. Its why we have lemon laws.

    (And this is why the notion of GPS driven cars scares me)

  • TastyBits Link

    If there is a change in the tactile feel, this could account for some driver error. People who have lost sensation in their feet & legs (diabetics) cannot tell the pedals by feel. They can only tell which is which through placement and feel. I believe they use a reference point, but if that reference point is lost, they could be pressing the brake (accelerator). There should also be opposite cases, but they may go unreported.

    If they removed the lead from the solder, this could explain the tin whiskers. It would be ironic if removing the lead for safety created another safety issue. Your government (EU) at work.

    “Then the bowsprit got mixed with the rudder sometimes:”

  • Drew Link

    Dave

    You can put what I know about software in a thimble. What type of error are you thinking?

    I had a Mercedes S 550 that would from time to time surge while coming t o a stop. Just a brief surge. No runaway. The dealers techs of course couldn’t find anything in the cars computer, nor could they duplicate it while test driving, but it was real.

    And so now I’m back with BMW.

  • TastyBits Link

    @Drew

    Tin Whiskers

    This was a possible cause of the Microsoft XBox “Ring of Death”. Re-fluxing could fix the problem, but it was a pain in the ass.

  • You can put what I know about software in a thimble. What type of error are you thinking?

    Any program beyond a certain level of complexity inevitably has bugs. Thinking otherwise is hubris. The figure I’ve seen quoted for the programming in Toyota’s onboard computer(s) is in the hundreds of thousands of lines. Clearly, that’s complex enough to meet the threshold I’m talking about.

    I realize that command and control software is the most straightforward programming there is. That doesn’t mean it’s going to be error-free.

    Modules interact in unforeseen ways; there are buffer overruns; memory areas that in theory are reserved for some particular use are used incorrectly for another; an unforeseen condition arises and a subroutine fails incorrectly. The possibilities are practically endless. I once knew a guy who chased what finally turned out to be a buffer overrun problem literally for months. Eventually I gave him a tool that enabled him to locate the problem. Man, was he frustrated.

    My understanding of the way they’ve approached looking at the software was to divvy it up among the members of a team of reviewers who, literally, read it line by line. I think the likelihood of learning anything useful that way is practically zero.

  • And so now I’m back with BMW.

    I drove BMWs for about 15 years, since shortly after I returned from Germany. IMO nothing matched the 2002.

  • Drew Link

    Dave

    Thanks. As for BMW, they ran through a period of quality issues. Those seem to have been resolved. The two best sedans I’ve ever owned were a 2002 540 and my current 740. The wife loves her x5.

  • It’s not just bugs in the software, but potentially weird interactions with all the vehicle’s sensors that supply data to the code. A code may run fine steady-state, but put in a couple of out-of-norm inputs and weird things can happen.

  • Icepick Link

    I don’t believe there is a metalurgist in me, I’ve looked.

    I’m only interested in metallurgy to the extent that sometimes it’s fun to watch a show on making samurai swords, and the occasional show on advanced manufacturing on TLC or the like. But that’s about it. It’s one of those fascinating topics in which I have zero interest. There’s a reason I like things like instrumental music, chess and pure mathematics….

  • TastyBits Link

    @Icepick

    I’m only interested in metallurgy to the extent that sometimes it’s fun to watch a show on making samurai swords, and the occasional show on advanced manufacturing on TLC or the like. But that’s about it. …

    Steel is to the America as concrete was to Rome. Steel plows allowed the Midwest soil to be worked. Steel beams allowed buildings to be built higher than a few stories. Steel cable (and Otis) allowed elevators to be used in the new buildings. Bridges, ships, etc. were all made bigger because of steel.

  • Icepick Link

    TB, I get that. It’s just that the subject doesn’t really interest me.

  • jan Link

    The two best sedans I’ve ever owned were a 2002 540 and my current 740. The wife loves her x5.

    A girlfriend of mine owns and drives a 2002 540, and gives it similar compliments. My husband has an X5 (which is our road car), and I have an X3. Except for the cost of repairs, they are great vehicles.

  • steve Link

    Why do BMW drivers keep getting ranked as the worst?

    http://www.independent.co.uk/life-style/motoring/motoring-news/bmw-owners-are-the-worst-drivers-1974852.html

    I love watching the samurai sword shows also. Friend who makes knives is planning on giving it a try. I have an invite for his first try. (I like knives a lot.)

    Steve

  • Drew Link

    Jan

    My current comes with maintenance included. They used to rip your lungs out. That’s why I went away from BMW. Can only imagine that they now include maintenance because they are confident in quality and know it will just be routine.

  • Icepick Link

    Why do BMW drivers keep getting ranked as the worst?

    My default position for years has been that people that drive German cars are assholes. This is based on their inclinations to cut people off, tail gate, flip off and cuss at other drivers, etc. I swear to God, when I was living in Maryland you could count on everyone that drove a German car to do all of those things to you, even if you were the only two cars on a five-wide highway in the middle of the night. People that drive bugs seem to be the one exception to this.

  • jan Link

    Icepick

    I used to think the same thing about high-priced German car owners, including being arrogant and cutting people off….until I owned one myself. I actually show extra ‘driver consideration’ when I’m on the roads, in a attempt to make up for the images cultivated by these car owners.

Leave a Comment