By on February 2, 2010

The media and “celebrities” are making hay over the Toyota recall issue, desperate to find evidence of electronic and software gremlins. We’re adamant in stating that Toyota needs to change their software to give braking priority over a stuck pedal, and to replace the pedals, of course. And there may well be genuine software or electronic glitches out there, but we’d like to see solid evidence of them. Instead, we’re stuck listening to Steve Wozniak’s experience with a faulty cruise control on his Prius. It’s being spun as an example of Toyota’s electronics gremlins, creating  confusion and scare-mongering. As if there wasn’t enough of that already.

Apple co-founder Steve Wozniak has described a repeatable situation that is clearly a flaw in his cruise control, and it only happens at super-legal speeds (from slashdot.org):

My 2010 Prius has a package that includes parallel parking assist and cruise control distance limiter. In some fictional state (let’s say ‘private property’) I tap my cruise control speed lever up and the car speed increases from 80 to 81.I tap this leve again and again, up to 83 mph. Then I tap it again and the car takes off without speed limiting. Tapping this lever down has no effect. The car is shortly up to maybe 97 mph. I repeated this many times.

One doesn’t think of things like putting the car in neutral instantly.

The natural braking action does disable this effect.

Wozniak has made this very clear: this only happens (repeatedly) above 83 mph. And a tap on the brakes releases the speed control, exactly as it should. This story is getting repeated play on Jalopnik, and other sites. Interestingly, Jalop modified their original from yesterday story to “sanitize” it, removing any reference to the illegal speed, as well as removing the link to his original comments (above) that describes the situation. I guess they though that might not go down so well. Instead, it’s being shown as evidence of Toyota’s callous disregard of their customer’s safety, because Toyota won’t drop everything and give the Great Woz a personal response to his complaint.

Well, the cruise control on my ’77 Dodge Chinook camper does the same thing, above 63 mph. Perhaps it’s trying to tell me something, because the one time I drove consistently faster than that I cracked an exhaust manifold. I’m digressing, but the point is this: Steve Wozniak’s problem is not a dangerous phantom or gremlin, but a consistent and repeatable flaw in his cruise control. And since cruise control is neither necessary and can be disabled, and the flaw happens only above 83 mph, perhaps Toyota legitimately has more pressing problems on its hands and feet.

Update : Jalopnik is reporting that thanks to its extensive coverage, Toyota will be taking Wozniak’s Prius for a week’s of testing. It helps to be a celebrity and be heard.

Get the latest TTAC e-Newsletter!

57 Comments on “Toyota Unintended Acceleration Gremlins Running Amok – In The Media And At Illegal Speeds...”


  • avatar
    ClutchCarGo

    [whistle]

    Piling on. Offense, Jalopnik. 15 yards, still second down.

  • avatar
    Philip Riegert

    So now all of the sudden everyone is coming out of the woodwork now. How coincidental.

    Like it was said before – isn’t any acceleration in a Prius unintended?

    • 0 avatar
      Robert.Walter

      Philip,
      This is not unusual … once one reaches a state or alarm or hightened-awareness and then begins an inspection one tends to find the thing being searched for as well as unexpected surprises … We saw this time and again when doing a quality control action for a certain flaw.
      Regards.

  • avatar
    Sinistermisterman

    Regardless of whether a fault occurs outside of legal speeds in the US, or is easy to ‘stop’ by touching the brake pedal, it is still another Toyota fault. There are places on earth (ie, German Autobahns) where these speeds are legal and a fault like this could come into play – People must remember that not everyone has the mental capacity to do the right thing when a fault occurs – remember that story of the Australian man and his Ford Explorer? Point in case.

    • 0 avatar
      Paul Niedermeyer

      People must remember that not everyone has the mental capacity to do the right thing when a fault occurs
      Like tapping the brakes?

    • 0 avatar
      Sinistermisterman

      You obviously still have faith in most people being even vaguely intelligent.
      Take this clever soul for example:
      http://www.telegraph.co.uk/news/worldnews/australiaandthepacific/australia/6824992/Speed-drivers-cruise-control-sticks-at-50mph-for-25-miles.html

    • 0 avatar
      cjpistonsfan

      ——–
      People must remember that not everyone has the mental capacity to do the right thing when a fault occurs.
      ——–

      They have padded rooms for people like this, not upholstered cars.

    • 0 avatar
      midelectric

      Interesting Telegraph link, not so much for the UA story but the vegemite sandwich picture in the Australia Day photo album perked me up a bit.

  • avatar
    SunnyvaleCA

    My car has an extra pedal that disengages the engine from the drive wheels. These devices are still popular in many parts of the world. Perhaps they’ll make a comeback here in the panic-stricken US of A.

    • 0 avatar
      chuckR

      90% of US car purchasers don’t share that hope.

    • 0 avatar
      psarhjinian

      Manual transmissions are not a panacea for sudden unintended acceleration.

      I’ve personally witness two people suffer “pedal confusion”** in a stick-shift car, and in both cases those people drove straight into the vehicle/object behind/in front of them.

      Drivers who panic and react badly will do so regardless of the car.

      ** most NHTSA investigations in to SUA and SLOC generally ended with a conclusion of “driver error”. If this Toyota escapade results in mandatory black-boxes in cars, I’d expect that statistic to increase.

  • avatar
    Mr Carpenter

    There is also that highly technical and difficult to master trick, called “putting the shift lever into neutral” which might be used by folks in situations like this. Assuming, of course, that they have two brain cells to rub together.

    Gosh, Who knew?

    • 0 avatar
      geozinger

      Until something like this happens to you, keep laughing.

      A ragged carpet remnant (I was using as a floor mat) got entwined at the bottom of my hopped-up Mustang’s gas pedal some years back, and when it happened I had no idea why my car kept accelerating. Frankly, I panicked, and stood on the service brake until the car stopped (the car had an automatic tranny). Once I was able to regain control of the car, I put it in neutral and disentangled the carpet remnant from the pedal.

      My issue was clearly my own fault, but I can easily understand why people may not have the presence of mind to use the ‘highly technical and difficult to master trick, called “putting the shift lever into neutral”’ maneuver.

      Believe me, it’s not so funny when it happens to you.

    • 0 avatar
      ihatetrees

      Until something like this happens to you, keep laughing.

      Yeah, those are check your shorts moments. I had a similar situation with an obstruction under the brake.

      I really don’t think these situations are all that rare. People need to know pedals can become stuck / obstructed.

    • 0 avatar
      Robert.Walter

      Geozinger: I agree with you . There is an old saying “pride and arrogance goeth before a fall,” as I’ve said in other posts, not every driver is a Sullenberger … it is incumbent on an OEM to design vehicles that do not assume this or test this.

    • 0 avatar
      Russycle

      I had a throttle cable on a Chevy 350 V8 get stuck with the carb wide open, I have to say shifting to neutral wasn’t the first thing that occurred to me. Fortunately it was pulling two tons of Suburban up a steep hill, and standing on the brakes I was able to come to a stop. In my defense I’d only had my license a few months, hopefully I’d react better after 30 years of driving. Actually, I think it would have been pretty easy to overshoot N and hit reverse, that would have been interesting.

    • 0 avatar
      fgbrault

      I’m sure there are many intelligent persons who simply have no interest in cars and view their car as just another appliance. Some of them, I’m sure, don’t know what “neutral” is, only that there is an N that they must move that lever past when going from P to D and vice-versa. They need to be protected.

    • 0 avatar
      cjpistonsfan

      —–
      I’m sure there are many intelligent persons who simply have no interest in cars and view their car as just another appliance. Some of them, I’m sure, don’t know what “neutral” is, only that there is an N that they must move that lever past when going from P to D and vice-versa. They need to be protected.
      —–
      You don’t have to complete educational courses and operational tests to legally operate a household appliance. You do in order to drive a car. Part of the education is supposed to include what different gears are and what they are for. Of course Toyota is not off the hook if their design is faulty, but it’s not the car companies’ job to make sure that people know HOW to drive a car.

      With that said, I’m sure that calmly discussing options in a discussion board and experiencing unwanted acceleration in real life are two entirely different things. However, as many on this website have evidenced through anecdotal history, those with the proper training who aren’t driving impaired ARE capable of reacting appropriately in those situations.

  • avatar
    jmo

    an example of Toyota’s electronics gremlins

    And somewhere a VW executive gets his wings. And you mocked me and my VW. I’ll have you know when my coil packs went there was hardly any acceleration at all.

    Harrump

  • avatar
    Mike66Chryslers

    The problem is obvious: If a Prius is accelerating above 83 MPH it must be going downhill.

  • avatar
    Guzzi

    The dude probably dropped a cheeseburger and it lodged under the pedal.

  • avatar
    thalter

    Doc Brown, from Back to the Future:
    “When this thing hits 83, you are going to see some serious shit!”

  • avatar
    ihatetrees

    Instead, it’s being shown as evidence of Toyota’s callous disregard of their customer’s safety, because Toyota won’t drop everything and give the Great Woz a personal response to his complaint.

    By the looks of things, Toyota’s attorneys should drop a few Big Mac coupons in the mail. Woz would call things even.

  • avatar
    Steven02

    All cars have defects. The Prius is no different. It is an interesting little defect, but not one as bad as the others Toyota currently has where the brakes don’t work.

    Honestly, I don’t think the speed at which the cruise control was on is that big of a deal. Consider that there are roads in the US with 80mph speed limits. For a short time, Montana had a “reasonable and prudent” speed limit. One would think that the speeds of the vehicle would effect cruise control like he is reporting it does.

  • avatar
    bolhuijo

    I have a car whose climate control system lets you adjust temp up to some number in the high 80s. When you reach that setting, it just goes into “full blast heat” mode. Maybe this was just a lazy programmer re-using some climate control code in the cruise control module. heh heh.

  • avatar
    Nicholas Weaver

    You guy’s aren’t getting Woz’s point here…

    This is clearly a software flaw, that he’s tried reporting multiple times, that has not gotten fixed. How many other gremlins may be lurking that aren’t so easily reproducable?

    • 0 avatar
      Aqua225

      Exactly. And can anyone elaborate on the Prius’s computer layout? Does the engine computer directly implement the cruise control, or is there another computer that watches the car’s speed and then requests more throttle directly to the car engine controller over the CAN network?

      Woz’s repro could *easily* be a function of the very same code causing unintended acceleration. Matter of fact, it could be the exact same set of conditions.

    • 0 avatar
      panzerfaust

      Why is it clearly a software flaw? I’ve had this sort of problem in cars that didn’t have a microchip in them much less software.

    • 0 avatar
      Paul Niedermeyer

      Nicolas, So he’s proposing that Toyota break the law to confirm his problem?
      Seriously, you have a point, up to a point. My point is that this seems like any other ordinary defect: the cruise control doesn’t hold a setting above 83; but there’s no intrinsic safety issue, is there? Sounds like a bit of code got lost or garbled. I suspect Toyota is prioritizing complaints these days.
      How many other software gremlins are there lurking? Impossible to answer, unless they rear their heads. But the existence of one doesn’t exactly prove that there are more.

    • 0 avatar
      Steven02

      Paul, no breaking of the law required. I assume Toyota has a test track somewhere a car can go 83mph. No breaking of the law required. Heck, put the thing on a dyno (now that would be funny to see, a Prius on a dyno).

      I think there is safety issue though. If your cruise control increment button decides that it wants to not increment but only accelerate, it probably isn’t safe. I am not saying that cruising at 83mph or higher is, but there are roads in the US where 80mph is the speed limit. And people do speed on all roads. Not too hard to see the use case.

  • avatar
    Aqua225

    panzerfaust:

    Simple, there is no mechanical or even *direct* electrical link between the throttle and the pedal on these cars.

    It is ALL gated by software. Even if the servo was getting stuck that controls the engine butterfly, the ECU should detect that fault as well, and shut down fuel & spark to safe the engine.

    Ie., when Toyota took the throttle into the digital domain, there became zero excuse for stuck throttle issues spinning out of control. The only way this could possibly happen, a scenario that the software could not eliminate, is if the pedal itself failed. Clearly, the pedal isn’t failing for the Woz, the entire throttle control is done in software either by two computers (a cruise and ECU) or one computer (if the cruise is part of the ECU software).

    • 0 avatar
      panzerfaust

      Point well taken. But the computer has to eventually translate its commands to an electronic throttle control unit of some sort. There still has to be a mechanical or electro-mechanical interface with the engine. Why couldn’t it be a problem there?

  • avatar
    panzerfaust

    Not knowing to put the shifter into neutral when the throttle is stuck, or the car has uncommanded acceleration may not be a of lack of intelligence; it may also mean poor drivers instruction. This is pretty basic stuff.

    • 0 avatar
      Aqua225

      Actually, according to some of the media reports, there are two possible scenarios about neutral:

      (1) The transmission is not disengaging under WOT.
      (2) Due to the wierd gate patterns most manufacturers are now using, locating neutral in a panic situation is not a straight-forward exercise.

    • 0 avatar
      Aqua225

      Perhaps the best solution is to install a “terminate ECU power” pin in all cars :) Pull the pin, and injector, spark, DBW servos are all disengaged. People would probably think such a feature to be cool anyway, and people could practice pulling the pin in empty parking lots. It could even be part of the licensing process: how to pull the pin.

      It would be better than the key, since it wouldn’t have the potential to lock the steering.

      I call patent rights, and this forum is my proof!

    • 0 avatar
      panzerfaust

      Sort of like the key on my treadmill? :)

    • 0 avatar
      sfenders

      I’m old enough to have learned to drive in a car with a cable-controlled throttle, which was somewhat problematic. After the first time it got stuck in unintended-acceleration mode, I made sure to train myself to be ready next time to instantly depress the clutch pedal and switch off the engine at the same time. No rev-limiter, you see. No power steering, either. If not for this little reminder from Toyota that similar things can happen even in a modern high-tech car, I’d probably have done the same now-wrong thing if it were to happen today.

      This Woz cruise control bug is somewhat alarming, really. I didn’t ever completely trust all this computerized crap anyway (I am a computer programmer), but would’ve expected them to at least catch all such reasonably obvious bugs in testing.

    • 0 avatar
      Greg Locock

      @sfenders- have you tried to put a cable operated trans into R at high forward speed? The reverse torque on the converter just stalls the engine, unless you apply full throttle at the same time, in which case it is quite funny in the right sort of car.

      In a shift by wire trans there is even less likelihood of a problem, it just won’t let you do it.

      Therefore there is little need to worry about ‘accidentally’ hitting R when you shift an auto into N.

      Incidentally on a more general note I tried combined braking and acceleration in 4 makes of cars yesterday, all allow it. T is not outside of industry practice in allowing it.

    • 0 avatar
      sfenders

      Greg, no, I never have tried shifting into reverse while moving forward at high speed. Somebody told me once that’s a bad idea. But yeah, somebody else once told me he used to have a car where you could be doing 20mph and put it successfully into reverse gear, that supposedly being a demonstration of how powerful the engine was.

      Personally I never had a car with automatic transmission, and with any luck never will. All the ones I’ve driven though, I think they had the little button you have to press to get to reverse.

      What I meant was that it’s wrong in more recent cars to try and switch off the engine right away, just because it’s no longer necessary, and losing power steering is not so good.

  • avatar
    Greg Locock

    So, has a reporter got into a 2010 Prius and tried to repeat the failure mode? Seems easy enough to do if you have access to Prii.

    If it is a routine sort of a software fault, eg that it regards any request for a demand speed above 83 mph as a request for top speed, then other 2010 cars with the same software should do it.

    If instead it is something odd special to his car then that is an entirely separate sort of a thing.

    • 0 avatar
      Paul Niedermeyer

      I’d bet a beer it’s a little glitch in his Prius only. There are way too many out there already, and plenty of Prius drivers actually drive quite briskly to not have had it happen to them.

    • 0 avatar
      sfenders

      You’d lose that bet on a technicality; it’s not a glitch at all methinks.

      I’ve just read Wozniak’s comment a bit more carefully. “I nudge it again and again, up to 83. Then I nudge it again and the car takes off, no speed limit.”

      Each nudge tells the cruise control to go a couple mph faster, and he did it “again and again” until he’d told it to go 120mph. Because he’s already at 80, air resistance is such that the normal amount of throttle the thing uses to speed up is not enough to do anything much. It takes a while for the cruise control to recognize this and go flat-out to reach the speed he told it to reach. In the mean time he gets impatient and presses the button a few more times. Silly Woz.

    • 0 avatar

      I think you are correct, sfenders. That makes the most sense to me. I still think it would be worth trying to reproduce the scenario on a test track.

      My question is, does the Prius have Decelerate and Cancel buttons? I would guess that in Woz’s scenario, Decelerate would have no effect (at least until it is pressed many times), while hitting Cancel would have the same effect as tapping the brake.

      Didn’t Al Gore’s son get ticketed for going ridiculously fast in a Prius? http://www.slate.com/id/2169925/

  • avatar
    Marty S

    The previous Lexus/Toyota unintended acceleration recall seems to have been eclipsed by the current one, but the previous one was more serious, with several deaths involved.

    I had an incident with my previous Lexus IS250 a year or so ago(now subject to the recall). I was proceding very slowly in a parking lot when the engine began to roar, revving very high, much higher than I would ever have depressed the gas pedal. I was not touching the gas pedal and tried to bring the car to a stop using the brakes. Even though I was going slowly, I could not overcome the force fo the engine with the brake sufficiently to keep the car from creeping forward. I put it into neutral, the engine revved out of control, but I was then able to shut it off. On restarting the car, everything was fine. The floor mat was anchored firmly in place and was not touching the accelerator pedal. I brought the car to the Lexus dealer the next day and they confirmed that the floor mat was not involved, and could find nothing wrong. A service order confirmed the incident and their findings.

    After hearing about the recall and the floor mat nonsense, I called Lexus customer assistance and told my story. The person on the line repeatedly asked me, “what would you like us to do” (since there was no damage and I no longer have the car – although I have a new one now). I told her that I would like Lexus/Toyota to be aware that such incidents have nothing to do with the floor mat issue, but really got no response. It seems to me that there is a problem in the electronics that they are not recognizing. Replacing floor mats and accelerator pedals in those models will not mean much. The other aspect of the recall, programming the electronics so that the accelerator is disabled when the brake pedal is depressed, is more important, and they are doing that, but not publicizing it too much.

  • avatar
    psarhjinian

    And since cruise control is neither necessary and can be disabled, and the flaw happens only above 83 mph, perhaps Toyota legitimately has more pressing problems on its hands and feet.

    It’s not so much that, it’s that you don’t say “Yup, you found a bug!” until you, the vendor, can reproduce it. Otherwise, you’re opening yourself to unreasonable liability.

    Woz should know this: if you report a bug to, say, Apple, they won’t acknowledge it until they’ve at least reproduced it and, unless it’s very serious, until they’ve fixed or at least worked around it. Even projects with open bug-trackers do not ack bugs until they’re reproduced.

    So that begs the question: there’s a lot of Pruises out there: can anyone reproduce this? Can they do it with other HSD hybrids? Anyone willing to break the speed limit in their Prius for the cause of science?

  • avatar
    mcs

    Now there are problems with the Prius brakes. Here we go again…

    http://www.marketwatch.com/story/toyotas-new-problem-prius-brake-complaints-2010-02-02?reflink=MW_news_stmp

    http://www.msnbc.msn.com/id/34611930

  • avatar

    Much like Audi in the 80′s (operator error due to pedal placement unlike the GM automatics they just sold) Toyota is going to be considered “an unsafe choice”. Unlike Audi, however, there is not a crew of auto enthusiasts who will soldier on. The target market for Toyota will run screaming, Consumer Reports (Toyotas now not “check rated”) in hand, to Honda or Nissan. Appliances don’t get enthusiasts. I recall the MR2, my family owned one. The Supra had some good years. Today, what ? The Camry Solara ? Yup, right over to that two door Accord, hold the dangerous horsepower, which Honda thoughtfully keeps away from the US market.

    I once had a throttle cable not return because a spring fell off. It was an old Firebird, and I turned off the ignition. No damage, but I was used to nursing that car along…..most folks don’t travel with a toolbox in the trunk, although today, you’d need a laptop too.

    Yet another example of a $15 dollar part which could have cost $16 and been reliable.

  • avatar
    John Horner

    Wozniak is, however, making a valid point. He has uncovered a clear bug in the electronics which control the Prius’ cruise control system. It really does raise questions about Toyota’s embedded software quality control. The underlying question is, what is the nature of the bug which causes this incorrect cruise control operation, and does a similar situation exist elsewhere in Toyota’s software? What is the cause? Unsensed buffer overflow? Runaway pointers? Incomplete state machine definitions? There are a whole lot of ways this kind of thing can go wrong, and there is a real chance that whatever software/hardware engineering error is manifesting itself as the runaway cruise control may also exist elsewhere in the code.

  • avatar
    RogerB34

    I’m more impressed that a Prius would attain 97 mph than Steve W’s cruise control issue. Too bad he couldn’t let it run for 5 min at full throttle.

  • avatar
    skor

    Everyone thinks that they could handle unexpected acceleration without breaking a sweat, so did I…..until it happened. I had the cruise set on a Caddy with the 4.9 while driving up the NY Thruway. Approaching my exit, I slid over to the exit lane, tapped the brake to cancel the cruise, and she kept going full speed ahead! I was half asleep after spending over an hour cruising in that floaty boat, but now I was wide awake. My first reaction was to stand on the brake. The brake scrubbed about half the speed then I dropped it into neutral. With the engine screaming I shut it off and coasted to the shoulder — smoke coming from the front brakes. Turned out to be a bad idle speed control motor. I shudder to think what could have happened if some little old in the early stages of senility had been behind the wheel.

    • 0 avatar
      stuart

      Well, I hadn’t really thought about unexpected acceleration when it happened to me in ’78 or ’79 (I would have been 18 or 19 years old). ‘Twas Dad’s ’73 Ambassador wagon, with a 360. Truthfully, it wasn’t full-throttle acceleration; I’d guess about 2/3rds (enough to open the secondaries). Happened on Hamilton Avenue, a flat, four-lane suburban boulevard (limit 35 MPH). When the car got over 50MPH, I tried lifting the accelerator pedal with my foot (no effect). Then I had the bright idea to disengage the transmission. About one-half second later, I had the brighter idea of turning the key to OFF. Note that the ’70s era GM column had a lock-out that precluded removing the key or locking the wheel unless the lever was in Park, but you could turn the ignition OFF. I did.

      Anyways, I coasted to the side, raised the hood, and immediately noticed the cruise-control bellows had come asunder. This was a ribbed rubber tube about four inches in diameter and length; when vacuum was applied, it would shrink in length, pulling a chain that opened the throttle. Inside the bellows was a large-but-soft spring. When the bellows popped off its moorings, the spring dragged the chain end cockeyed, far past the usual travel, pulling the throttle open.

      I stuff the spring back inside, reinstalled the bellows, and it never gave us trouble again.

      I commuted to college in that car for a time; the trip took me over the Santa Cruz mountains. I routinely used “Mexican overdrive” on the downhill part (approx 8 miles) of the trip to save gas. Fun times.

  • avatar
    pete darnell

    This happened to me on rt 84 in CT a few years back. I was driving a ’98 4 cyl dodge Caravan w/120K miles. Fortunately it was on an uphill and the car was so under powered that it didn’t get much past 85 mph. What unstuck it was to stomp the gas pedal a few times to loosen up the throttle mechanism. I blasted the cable and return spring mechanism with some lithium grease and it never happened again (well, at least not for the remaining 10K miles I got out of it before the engine blew).


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

  • Brendan McAleer, Canada
  • Marcelo De Vasconcellos, Brazil
  • Matthias Gasnier, Australia
  • W. Christian 'Mental' Ward, Abu Dhabi
  • Mark Stevenson, Canada
  • Faisal Ali Khan, India