Slashdot Mirror
Car With A Mind Of Its Own -- Part 2
An anonymous reader writes "As a sequel to the previous Slashdot story where a car 'began accelerating to 120 mph on its own', Renault (the car manufacturer) has examined the supposed faulty car, and as many of us have suspected, no anomaly has been found (google translation). Renault will initiate a court action to discover the truth about the matter. Read more about it here (translation)."
I believe it is all drive by wire. BMW steering and throttle is all by wire. Not sure about brakes. But considering the fact that most every car has power assist brakes.. it wouldn't surprise me.
She escaped with only burned out brake calipers.
The fault was a bit of grit or buildup preventing the throttle from closing properly.
Keep your air filter clean and don't buy junk gas.
If you remember that Airbus that crashed at an airshow a few years back when it's Die-By-Wire flight-controls refused to give the pilot TOGA power.
That accident was put down to pilot-error by Airbus and the French (Government) Investigators. The case has now been re-opened on the merit that the CVR and FDR data seems to have been played with.
Ripping an new rectum in the fabric of spacetime.
I seem to remember reading an AutoWeek article about 2 years ago about the time that DaimlerChrysler's Jeep divison introduced the Jeep Liberty small SUV.
1
3
I opened my mail, and I saw this full page cover photo on this weekly auto magazine showing a Jeep Liberty tipping over during a slalom test. An inset picture showed the friggin' car flipped over on its side, if I remember correctly. I'm posting from work, so read for yourself:
http://www.autoweek.com/article.cms?articleId=333
Turns out, AutoWeek testers were doing their standard lane change avoidance/slalom test that they do with everything from Hyundais to GMC Yukons. I'm pretty sure it was a production Jeep Liberty -- nothing pre-production -- that flipped over twice (???) and landed on its side during this relatively commonplace automotive review test.
The driver, thankfully, only suffered a sore neck (nearly broke it, if it had rolled one more time), and AutoWeek devoted their entire issue to this vehicle which had been designed to put an affordable small SUV Jeep into the hands of consumers.
DaimlerChrysler balked and basically claimed that THE TEST WAS NOT A REAL WORLD TEST. AutoWeek called bullshit and basically said, "Uh, yeah it is -- if a driver has to make a quick lane change and or dodge something in the road, it's as real world as it gets."
http://www.autoweek.com/article.cms?articleId=416
I seem to remember that DaimlerChrysler continued to balk at the test, but in fact they ended up making center of gravity changes to the vehicle (suspension and ride height, perhaps?) over the course of the next model year.
Sounds like the same crap that Renault is doing here.
It's funny -- the automotive press gets touted all the time when they LOVE a car and try and hype up the manufacturers' products, but heaven forbid that they also try and save the manufacturer a little legal trouble by finding out these sorts of dangerous rollover issues and what not in pre-production cars. Only the GOOD NEWS, right? Bullshit.
IronChefMorimoto
Brakes of a car can always overpower an engine, even at full throttle.
Very Very wrong.
breaking prolonged and at high speed will cause severe reduction in breaking power and even a LOSS of breaking power.
The mechanics of this decline and failure in the coefficient of friction are varied. At a certain temperature, certain elements of the pad can melt or smear causing a lubrication effect, this is the classic glazed brake pad. Usually the organic binder resin starts to go first, then even the metallic elements of the friction material can start to melt. At really high temperatures the friction material starts to vaporize and the pad can sort of hydroplane on a boundary layer of vaporized metal and friction material which acts like a lubricant.
so brakes not working in this instance is certianly a possibility.
and yes I said this last time. It amazes me how many people really do not realize how brakes work.
the only solution to avoid the above is 3 piston calipers and vented, finned and crossdrilled rotors. Even the most expensive cars do NOT have these performance features. The only car's I have seen them on were as an OPTION on Ferarri and the Corvette ZR1.
Do not look at laser with remaining good eye.
The Cruise control buttons (on, set, and resume) we beneath the 5 speed shifter and the computer remembered the last speed used even after the car was turned off and on.
One day, I decided to play a trick on my wife, because she hated the car. In the morning, I set the cruise at 110 mph. Later that day, we were on the Turnpike driving somewhere (she was driving) and I casually hit the resume button and the car acclerated at full throttle from 75-80 mph to over 100, with her freaking out the whole time.
I was actually quite surprised how much she freaked out. She's a very intellegent person and all she had to do was step on the brake or clutch, but she just flipped out as it accelerated.
On the way home I did the same thing and her reponse: to pull the car into neutral. I watched the engine spin to 9k rpms (quite rapidly). Not so good. At which point I freaked out and told her about the "trick."
In summary, I was very suprised that someone as intellegent as my wife completely lost it as the car accelerated, so I can easily see how others would react in this situation. Hiiting the breaks/clutch/etc... may not be easy for someone to comprehend at the time of unwanted acceleration.
That model probably hasn't the usual lever but two buttons on the steering wheel, and, you guessed it, the gearbox is electrically actuated with an electronic control.
BTW, the hand brake is also electrically actuated. I know it because on a Renault Scenic I got stuck in the rush hour because the main fuse blew. Of course the manual override is in the boot, and the boot has no mechanical way to open it, only elecrical, so it was impossible to open.
BTW2 the speed control has at least a bug: while the nominal increment is 2Km/h (i.e. you can preset 30, 32, 34 and so on) under some circumstances (i.e. not always) it wouldn't let you set the speed at 50Km/h: it's either 48 or 52.
Due to these facts I tend to believe the guy more than Renault.
Back in the mid 1980's just before the Audi 5000 thing was in the news, a schoolmate's father bought one of those cars and was able to repeat the symptom by engaging the cruise control, then lightly double-tapping the brake pedal, which is supposed to disengage the cruise. The accelerator pedal would suddenly drop to the floor and the throttle would go wide open due to the cruise control servo pulling it and not releasing even if you turned tried to turn off the cruise. The brakes were not very effective at WOT since the engine was quite strong. Placing the transmission into neutral or turning off the ignition switch was the only way to stop it. He took the car back to the dealer to be serviced and two days later he got a call to come down to the dealership, thinking that the car had been repaired. When he got there, there were two suit-wearing bigshots from Audi, who offered him a check for the full purchase price he paid for the vehicle. He signed some kind of non-disclosure agreement, took the check and bought a different brand of car.
Yeah I'm not ready to call this guy a liar yet either.
Car's are increasingly drive by wire nowadays. My 2004 Mazda3 has 100% electronic throttle control, there is no mechanical link between the gas pedal and the throttle body. Also my power steering fails if the engine isn't running, even if the engine is still turning (i.e. coasing down a long hill in gear with the ignition off).
Some new luxury cars will start and drive with the ignition key only in the proximity of the ignition, this Renault may have that feature as well. Throw in an electrnically controlled clutchless transmission, wither it be SMG, CVT or electronic auto and I'm willing to entertain that his car could have had the Renault equivalent of a BSOD and sent the throttle to 100% to stay.
Now if there really isn't signs of significant wear on the brakes that certianly discredits his story. Hitting the brakes hard at 120 would definitely do some damage to the pads/rotors. Or maybe he just didn't have the balls to stomp on the brakes when they weren't working as well as he wanted.
considering that in a frontal collision, the steering column hitting the driver is a common cause of death, you have to wonder why this is a requirement...
I had the exact opposite problem happem in a 300ZX. My mechanic had done some work but forgot to tighten the bolts holding the throttle cable down. It came loose on the interstate, causing the car to drop back to an idle. I hopped out, took a look, and put the throttle cable back where it was supposed to go and hand-tightened the bolts.
These days I know that pretty much any vehicle with a mechanical throttling system is not 100% trustworthy, but I also know how to deal with situations where the linkages get stuck so it doesn't tend to be particularly inconvenient when it happens.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
> Shift into neutral.
You forgot to suggest trying the brakes first.
According to the horribly mangled translation of the article, it appears that Renault examined the car and found no wear on the brakes. Assuming they don't find anything that would prevent the brakes from working, that suggests the driver didn't really try to use them to stop the car.
I wonder where runaway acceleration is more likely: a new car or an older car.
My own runaway acceleration story:
I used to drive an old beat up VW Rabbit diesel. It was my dad's car which he let me use while I was in college. It used to accelerate uncontrollably after driving for 20-30 minutes. It only happened during highway driving.
There was nothing visibly wrong with the throttle cable. Immediately after stopping (by applying brakes), trying to start the engine would immediately result in the same problem. If it was allowed to rest for about 20 minutes, it would drive OK again. Dad's analysis was that my big feet were just getting caught (thanks, dad).
When it started accelerating, the only way to stop it was to pull over and push down on the brakes until the engine stalled.
It took about a year before we found a mechanic that knew the answer. There was an exaust line which would allow motor oil to blow into the engine. The motor oil/diesel fuel mixture would burn hotter and faster than just the diesel fuel, causing the engine to race. We pulled the exhaust line and the problem went away.
The clouds of smoke pouring out from the hood caused it to look like the car was on fire every time I stopped at an intersection, but it never took off again.
The reason we went to that mechanic and finally discovered the truth is once while on the highway it *really* took off. There was less time to think than usual because of heavier traffic. I forgot myself and hit the clutch. That did stop the engine, but in a far more spectacular way than I would have preferred (motor oil spraying out of the hood).
The car also used to cough big globs of used motor oil out the exhaust pipe. Nobody ever used to tailgate me.
breaking prolonged and at high speed will cause severe reduction in breaking power and even a LOSS of breaking power.
Well, there's the rub, though, isn't it? If a car goes full throttle and out of control, who's going to let that continue for a prolonged period? A sensible person would shove the brake pedal to the floor and immediately bring the car to a stop rather than continue tooling down the highway applying partial braking force until the brakes overheated.
So, the poster was very, very right.
- Hail to our fearless misleader! Fool speed ahead!
I disagree. If the guy had been fighting the vehicle for control, it would show forensic evidence of the attempts: condition of brakes being an obvious one, if they'd been applied hard while the car had an open throttle and was hitting 200kph. I'm not aware of any car in which every system that could bring it to a halt is run from software.
When I lived in Phoenix, there was a woman who had her Hyundai accelerate out of control. She was blowing through red lights, barely managing to keep her car under control. The police were trying to clear traffic ahead of her, and finally one got in front of her and used his brakes to slow her car down. It was a frightening event, and everyone was just happy that the woman was safe, and that she hadn't killed anyone else.
And then, it happened again. Turns out she was just an attention whore, and nothing was wrong with her Hyundai.
Fascism trolls keeping me up every night. When I starts a preachin', he HITS ME WITH HIS REICH!
Comment removed based on user account deletion
A) They could have a bug that only comes out under particular cirumstances that weren't reached in the test.
We all know that.
B) It can take a long time to smoke out weird car problems. I had a problem with my steering lock engaging that wasn't correctly diagnosed for like two years because it was a very specific set of actions that caused it to happen. It eventually caused an accident; fortunately it just put me into a guard rail when I was making a turn. They didn't believe me until one of the mechanics was buzzing around the shop and it almost put him into the wall.
THEN they believed me.
The revolution will NOT be televised.
True, but if you held the brakes on (Say for a minute) and noticed absolutely no change in vehicle speed - Would you even bother to try any more? I know I wouldn't - I'd be more interested in concentrating on the road ahead / finding other options to make sure I didn't waste someone. This could of course explain why his brakes weren't left as smoking metal strips at the end of it.
I'm getting the feeling with this one that we'll never really know what happened.
People that believe in their opinions don't post AC.
Except that there is STILL a mechanical component in throttle by wire, an dthe mechanical component is suppose to fail closed. Throttle butterflies (and my diesel pump, are spring loaded, and the actuator pulls against the spring.
Yes, it is possible for the actuator to jam open instead of just stop working - all systems fail. But the strong tendency is to fail closed.
Contrast this with steering. With steering, there is no "safe" fail position. Left, right, straightforward - it's all a pretty bad choice.
Braking is a different problem. Cars don't have "fail-safe" brakes - that is, if there is a brake failure, the brakes are applied, not released. Trains, and I think trucks (anything with air brakes) have them, but not cars.
Also, we need to define the term "by wire" here. I, and most others, use that term to describe a system where a physical linkage (mechanical or hydraulic) is replaced by signals, a processor, and an actuator. So far, NONE of the "counter examples" in this thread for brakes and steering qualify. Those systems are electronically assisted, where the driver still supplies the motive force, but that force is modulated by the processor. Take the processor out of the loop, the driver is stil steering or braking. Yes, those systems are more unreliable, and yes, they fail "harder" than strict manual systems, but that is a world away from "drive by wire."
Is it coming? Yes. I saw an article on an experimental joystick controlled Saab a few years ago. But it will be a while until I trust a safety system that does not have some kind of backup. If the backups fail on my current car, i.e. I fail to provide the proper foot pressure or steering force, that is my own damned fault.
"As God is my witness, I thought turkeys could fly." A. Carlson
Not related at all to what's going on. My wife was with me, and thinks it's a funny story.
"Learning is not compulsory... neither is survival."
--Dr.W.Edwards Deming
How many people actually know someone that is a competent driver that has had this happen?
It's happened to me before. One winter while I was still in Anchorage the throttle to my car would jam wide open for no apparent reason, usually while I was going 40mph+ Over the course of about two weeks I had this happen about half a dozen times, every time I had to kill the ignition and pull over. Power brakes and steering typically stay pressurized long enough after killing the engine to get the car safely stopped at the side of the road.
As to the cause of the problem. The protective plastic sheath that was wrapped around the throttle cable had developed a lengthwise crack and water got inside. What was happening was that I'd start the car and let it warm up at idle for 10-15 minutes before driving. That was more than enough time to thaw out the water in the cable but once I got on the road there was enough airflow to refreeze the cable, with the throttle wide open. Very scary. Took my mechanic a while to figure it out because he would let the car warm up in his garage for a while before working on it. It was cheap and easy to fix once he figured it out but made for a scary couple weeks of driving!
"Listen: We are here on Earth to fart around. Don't let anybody tell you any different!" - Kurt Vonnegut
I don't think he could just put the car in neutral. Some cars actually won't let you do things that would hurt their engines or transmissions.
Many manual transmissions can't be put into reverse if your travelling more than 10 mph in the forward direction for example. My Audi won't let me shift out of neutral and into reverse without requiring me to first put my foot on the brake. It's also got a locking mechanism so that if you're in park, you can't shift without pressing the brake pedal. I have heard of Audi owners having problems with this... one guy couldn't get his car to go into any of the non-forward-gear settings.
The transmission has a little computer-controlled actuator that does all that stuff.
And I had major engine work done on my car a while back. When I got the car back from the shop, the cruise control was screwed up; it wouldn't activate, then all of a sudden it would set itself and go. It was kind of neat until it got stuck on "accelerate". It got up to 115 before I got it to shut off by braking hard. Fortunately, I got it fixed (replaced a computer module that was shorted when they worked on the car previously), and I was out in the high desert of California when it happened (no one lives out there).
And I've seen cars whose throttle got stuck in the full open position, and whose drivers put the brakes on until they caught fire (local news video... quite a few years back). So stuff like this is fairly believable.
I was driving along on a highway on a Sunday morning when my main ABS unit with all the actuators, solenoids, and such blew a seal. Since it ties to all the brake lines, there was a sudden complete loss in brake pressure. I was stuck with a car going 55mph and a brake pedal that sank straight to the floor. Having an automatic, it wasn't easy for me immediately downshift, but I got it and was able to slow down enough to use the cable-actuated parking brake to come to a stop. Needless to say, the drive to the garage was a slow one.