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)."
No anomaly found? Of course not. This guy is full of shit, plain and simple. A similar problem almost put Audi out of business in the 80's because of a "story" on 60 Minutes. These people were just as full of shit as this guy.
? artid=22184&pg=1
How many people actually know someone that is a competent driver that has had this happen?
Linky to the Audi story here (google's first result): http://www.auto123.com/en/info/news/news,view.spy
Anyone remember the Audi disputes in the 80's where people kept claiming they randomly accelerated when the brakes were appled? I think it turned out that the accelerator and brake were too close together and people were hitting the gas pedal instead of the brake.
This way to the egress...
This is a fancy way of saying "The guy is lying".
You were mistaken. Which is odd, since memory shouldn't be a problem for you
Yep:
KITT = Knight Industries Two Thousand
KARR = Knight Automatic Roving Robot
So I watched a lot of TV as a kid. So sue me. (Not that I'm worth anything.)
Javascript + Nintendo DSi = DSiCade
...the power of one lying bastard when given media attention. Once upon a time, when I was young and foolish, my best friend and I contrieved a scheme to get out of a speeding ticket. We figured that if we were caught roaring past a speed trap, we could just continue, and call 911 to get them to clear the road. Then we could get to a safe area, spin out in a huge cloud of dust and praise be to god, we'd be OK. Of course, this would hinge on us being able to convice the cops and investigators, that we weren't actually maniacs. Like I said, when I was young and foolish.
Much harder with an automatic gear. The thing is, nobody expected Renault to say: "Oh yes, sorry our car almost killed you".
The guy was panicking. I'm sure that I would have reacted the same way. Do you realise that he called the cops himself? How on earth could you do that if you were just making up a bad joke or trying to get home sooner (which is total bullshit, since he missed his exit and drove over 50 miles away from his home...)
Anyway I believe this guy and we should wait for the real experts conclusions.
--
kTag
She didn't have a shift lever to downshift or put into neutral, or a key to turn off?
Carbon buildup on the throttle body is pretty common, but you usually get a lot of warning; a slightly sticky (usually sticks closed) throttle for several months/years before it gets that bad.
FYI, from their media site:
After one of its vehicles is incriminated
RENAULT TAKES LEGAL ACTION
Under the control of an independent court-authorized expert appointed by Renault and with the
consent of the customer, Renault evaluated the Renault Vel Satis 3L Dci automatic, registration
number 218 TH 18, in static and dynamic conditions on Wednesday October 6.
According to the driver's comments as reported in the media, the car was jammed at high speed
on the A71 motorway on Sunday October 3, due to a faulty cruise control. The driver reportedly
said that it was impossible for him to stop the car after trying different possibilities for almost an
hour.
The evaluation covered all the vehicle's electronic, mechanical and hydraulic functions. The
data collected and the facts as established reveal no malfunction. The braking system, which
shows no sign of abnormal wear, the gearbox control, the powertrain and the cruise control all
worked perfectly. A full detailed report is expected very shortly.
Given the findings of the evaluation and its concern about the impact this incident might have on
its product image, Renault has decided to take legal action in the form of a summary
proceeding, without prejudice to other actions taken in compensation for any damage suffered
by the company.
It used to be on the written drivers test where I lived that the correct response to this was to "kick" the gas pedal to dislodge it or to shift into neutral on an auto or use the clutch on a stick (either will probably blow the engine). Turning the key off was not preferred as it locks the steering wheel on most cars causing a loss of control.
Your statement has the same weight as Irwin Fletcher saying "Its all ball bearings these days."
Throttle by wire - a couple of companies
Brake by wire - none
Steering by wire - none
Why? If the first one fails, the engine dies. If the second 2 fail, you die. Some carmakers are experimenting with it, but none have brought to production.
"As God is my witness, I thought turkeys could fly." A. Carlson
Actually, BMW has resisted going to steering by wire. BMW is big on being the "Ultimate Driver's Car", and wants to keep "road feel" in the steering wheel. As such, there will be a mechanical link between the steering wheel and the tires for quite some time to come.
They have had a reputation for making outrageous cars like this..
They got a featured article for it as well!
IF he had used his hydraulic brakes as hard as he could, and the engine was still being told to go WOT (wide open throttle), there is some chance the car could still keep going forward, but not before the calipers and rotors would be pushed beyond brake fluid boiling point, and lose his brakes entirely. Even so, it certainly would have slowed him down considerably.
. php&carnum=626
From pictures I found on the web, this thing has an automatic transmission, and a lever to control it (like most cars). Neutral could probably have been used. The engine might have blown, but he would have stopped.
From reading, this car does have a motorized throttle valve, which could be a culprit. It would still require dodgy panic thinking on the part of the driver not to realize his options and/or he is full it.
http://www.ultimatecarpage.com/frame.php?file=car
Driving at that speed is not simply "reckless driving" in France. A few years ago it was upgraded from a simple offense to the crime of endangering other road users.
... speeding and driving under the influence.
Also, according to other articles (can't be bothered to find the URLs), the guy had just gotten his license back after having it suspended for
Maybe I should back up and tell more of the tale.
By first instinct, she hit the brakes (she was doing city driving), which then burnt. She then figured out what was going on and shifted into park and killed the engine. She didn't go all that far after she discovered what was going on (brakes were due to be changed anyway).
I never saw the symptoms, and the air filter was due for a change, so the bet was something got through the air filter.
Floor accelerator, then release to free throttle cable (won't work on throttle by wire, usually). If this doesn't work ...
Shift into neutral.
Apply service brake (or parking brake if service brakes fail) and GENTLY stop the car.
Do NOT turn off the key until you have stopped moving, as this will cause the steering column to lock.
Fortunately, I have only had my throttle stick once, about halfway, and I didn't have to go past step one. It was very cold weather and I imagine it could have been some ice in just the wrong spot.
Gamingmuseum.com: Give your 3D accelerator a rest.
"Renault decided to take action in justice in the form of a procedure of summary procedure which will lead in particular to a contradictory expertise, without damage of other actions in compensation for the damages that the company would have undergone", according to an official statement of the manufacturer.
Maybe they need a different spokesperson...
There's definitely truth to the phrase "lost in translation"...
"Truth is not decided by majority vote" consensus gentium -- Norman Geisler
- the driver was not drunk;
- he actually called the police while driving at 200 Km/h to ask for help;
- he had not been ticketed before the call or caught by any police radar that would have made him want to pull that stunt to cover his speeding excess;
- he eventually managed to disconnect controls to the car so that it finally and slowly cruised to a stop on the emergency lane.
Yes, the driver might still be lying about the all thing, but, if so, it was not to cover any known misdoing. On the contrary.Well the ignition is by wire usually, and most cars turn off when you shut off the engine :-P
Throttle by wire - a couple of companies
Just to give an example, it may be more than just "a couple of companies".
I drive a compact VW with a 1-liter, 16-valve engine, and it is "throttle by wire". That's an economy car, mind you.
I had if fail on me a few times when new, the dealer would never figure out the problem, they would see some strange conditions logged on the CPU, hard-reset the whole thing and it would not happen for a few days.
After the third visit to the shop they finally replaced the whole control unit and I never had another problem with that car since (had it for 2.5 years not)
... nearly ruined Audi in the 80s, thanks to a few issues with the SWI*..
Turns out folks were ham-footin' the small brake pedal and hittin' the gas at the same time with their big fat American feet. That's why automatic cars have shift locks now. OTOH, you will never hear the FedGov call a voter/taxpayer what they really need to be called: a dumbass. P. J. O'Rourke wrote up a really good commentary on this in one of his books that I can't recall right now, but it was pretty spot-on and funny, as most of his stuff is.
* SWI: Seat to Wheel Interface.
Your comment reminds me of the minority report where the one false positive is considered by the public to completely outweigh all the good the system has done.
I cannot speak for the incident you are speaking of, or even of flight controls at all, but I do know that mercedes was able to reduce its accident rate by 1/3 in Germany by introducing a "drive-by-wire" stability management system on some of its models. Maybe this system will in some cases make mistakes. But a 33% reduction is a pretty drastic reduction- one that will make me sleep better at night, and is an acceptable risk for me.
I personally am looking forward to the day when my car will take control and brake for me when I have momentarily taken my attention off of the road and the guy in front of me has slammed on his brakes, or alert me that I am going off the road if I am falling asleep at the wheel. And even just vanilla traction control is an absolute godsend when driving my Mustang through the snow.
Some of these systems seem so easy to implement too- I mean how hard can it be to put a sensor in front of the car and sense if you are approaching an object at a speed that you will soon not be able to avoid a collision? I understand it is not quite *that* simplistic, but surely it can not be much harder than implementing an airbag system. I am guessing it is mostly the fear of litigation that keeps such systems out of cars today.
ABS is essentially brake-by-wire. If the black box decides it doesn't want pressure at the slave cylinders, the driver is just along for the ride.
I do not know about other ABS systems, but in my vehicles if the ABS fails the power brakes still work. I had my ABS malfunction in one vehicle and the brakes worked like normal. In my experience, ABS modules are very fail-safe and I have never heard of one taking the driver along for the ride. Maybe you have, I am not saying you are wrong, just that I have not heard of that before.
24 beers in a case, 24 hours in a day. Coincidence? I think not!
Try taking a look underneath a Z4 sometime. "Electric power steering" refers only to the system that is providing the BOOST to the steering effort. The actual connection between the steering wheel and the tie rods that turn the wheels is EXACTLY THE SAME as any other mechanical steering system...either a recirculating ball or rack-n-pinion system.
The ONLY difference here is that the boost is provided electrically rather than hydraulically. Your example is "steering boost by wire", NOT "steer by wire".
Could someone please point out to me where in the Constitution, exactly, is the "Right To Not Be Offended"?
Turning the key off was not preferred as it locks the steering wheel on most cars causing a loss of control.
I drove stick cars for many years. My first was a beetle and my latest a Mustang. I had a throttle stick once (broken spring) and I didn't want to blow the engine. Plan ahead on what to do. Pratice it. I calmly turned the key one click only. Problem solved. Nothing broken.
Newer automatic cars are even simpler. Turn the key off. It won't go into lock until you put the shift in park. They do that on new cars so they won't roll away if you forget to put it in park as it keeps you from removing the keys. To get your keys out, you have to put it in park and then lock the steering. Only then can the keys be removed.
Try it. Shut off the car in your driveway but leave it in nutral or in drive. Try to lock the steering and remove your key.. There are only a few cars that let you remove your key without locking the transmission and steering.
The truth shall set you free!
One month ago BMW called back worldwide 580 X5 3.0 with manual transmission.
l _i d=6924
When touching the clutch slightly, the car accelerates at full throttle, even is the gas pedal is pushed only a little.
http://www.autobild.de/suche/artikel.php?artike
The term is "Could NOT care less"! Saying you "Could care less" implies that you do indeed care.
The phrase "I could care less" is sarcasm. As in, "I could care less... if I tried really hard, but it's not worth the effort".
If you had read the other story, you read that:
Selon LCI, le conducteur avait déjà été condamné pour état d'ivresse et excès de vitesse, son permis lui avait été retiré durant 4 ans.
Translation:According to LCI [a TV Channel], the driver has already been prosecuted for drunken driving and over-speeding, and his license cancelled for 4 years
This guy sounds like a dangerous idiot who is trying to protect his but with a fake story.
It also sounds like the media picked up HIS story first without fact-checking. Same thing happened with a woman in the Paris subway who claimed she and her baby were attacked by "anti-semites". Her story went first page in most of the newspapers, people started shouting against the insecurity in France, "anti-semitisme" and all sort of non-sense. It turned out she was mythomaniac and made up the whole story.
Actually, it's not the sudden opening of the throttle that would result in unburned fuel.
In the old school carb, if the throttle (a simple metal plate) opens wide rapidly, it will result in greatly increased airflow without much of an increase in fuel flow - and the engine will tend to 'lean cut' - not enough fuel and far too much air for combustion.
The solution to that was to add accelerator pumps to the carb. When the driver boots the throttle, the mechanical linkage also activates one or more accelerator pumps which squirt a bunch of additional fuel into the mix. Naturally, this isn't metered particularly accurately, but it'll ensure that the mixture is more or less right (probably well on the rich side) so the engine doesn't lean cut.
Modern electronic fuel injection is a completely different kettle of fish. Cars without pure throttle by wire will still meet requirements because the EFI system won't just dump a bunch of raw fuel into the incoming air like the accelerator pump, it'll be metered. In many cars, the linkage to the throttle butterfly is still just a cable - but pure throttle by wire (where the only connection between your right foot and the throttle body) is better still as when the engine control computer (known as a FADEC - Full Authority Digital Engine Control - in aviation) has control over the entire process instead of reacting to the throttle butterfly suddenly going wide open.
Oolite: Elite-like game. For Mac, Linux and Windows
http://sunnyday.mit.edu/papers/therac.pdf is the work on the Therac-25 problems by Nancy Leveson. If you have ever worked with software for embedded/control/fault tolerant systems you know her name. She pretty much invented the area of Software Risk Analysis. I don't doubt her explanation of the issue. Maybe Renault needs to hire her to find out IF or WHAT really happened. IIRC, she is at MIT now.
They still have a mechanic backup, because here in germany no car with steering by wire or brake by wire is allowed on public streets...
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
I've had my cruise control accelerate beyond its setting several times. Even had the logic module replaced twice to no effect.
I'll have braked to slow down for other traffic, then passed that traffic, and hit Resume. The car accelerates to the previously set cruising speed as normal, but then suddenly decides it has to go floor it to reach another, higher speed.
Tapping the brake pedal once has always disengaged it, and the one time when I had enough clear straight road to allow it to run (the technician that worked on it wanted the data), it would go up to 95 MPH, slow to 90, then back to 95, repeating. Probably a limitation in the vehicle that prevents it from going faster.
Unfortunately I have been unable to reproduce the behavior on demand, and I always wonder if it is going to go crazy again the next time I hit the Resume button. I'm thinking it must be some combination of the cruise controls used to adjust the cruising speed pressed long before the triggering event that primes the event.
The only way I'd accept a black box in my car would be to diagnose this problem and get it fixed, and then I'd have the black box removed.
Mine is a Honda Civic with aftermarket cruise control (not a standard option).
Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
Actually, that's completely incorrect.
The aircraft did give the pilot TOGA power (take off/go around power), but the aircraft was ALREADY so far behind the "power curve" (i.e. in the region of the flight envelope where slowing down actually induces MORE drag - if you want to find out more google for a drag/airspeed diagram - unlike ground vehicles, the curve has a "back side" for aircraft where induced drag increases as speed decreases) that it was simply against the laws of physics for the pilot to extricate himself from the pickle he'd put himself in.
It takes several seconds for a turbofan engine to "spool up". Unlike a small piston engine which can go from idle to maximum rated power almost instantly, a turbofan engine takes several seconds to go from a low power setting to takeoff power. There's an awful lot of inertia in the many turbine and compressor discs. (Modern jet engines are much better than the first generation ones, but they still take time to get up to speed).
The pilot did ask for takeoff thrust. However, by the time the engines did reach takeoff thrust (they actually performed slightly better than spec) the tail of the aircraft - which was already in a nose-high attitude because of the angle of attack needed to fly as slowly as he was flying - was already striking the trees at the end of the airfield. The additional drag of pulling the empennage through the trees overcame the thrust of the engines, and the plane slowed further causing it to impact more trees, adding more drag, slowing the plane further, until the final impact with terrain.
The pilot was ENTIRELY at fault. The same thing would have happened if he was flying a Boeing 727 ('Jurassic Jet') with the same kind of flight profile. He tried to violate the laws of physics and lost. The overall record of the Airbus A320 series shows that calling it "die by wire" is pure, unadulterated inaccurate hyperbole.
Oolite: Elite-like game. For Mac, Linux and Windows
Didn't Daimler Chrysler learn anything from the famous Moose Test ? This is where some Swedish journalists easily put an A-class on it's side in their standard Moose (called an Elk there) avoidance test.
Chip H.
with a mechnical cruise control. I worked with Ford investigating "run-awy" vehicles in the early 80's. Turns out there was a real problem. There was an aluminum linkage that rubbed against a guide such that it could eventually wear a notch in the linkage. The notch could get large enoguh to eventually catch on the guide, and the cruise control was stuck. No way to turn it off other than shut down the engine. However, the notch would form at the location corresponding to your mostly commonly driven speed, so say here in Houston you would get the cruise typically stuck at about 90 mph.
----- There are two kinds of people in this world, my friend; those with loaded guns, and those who dig.
http://www.edmunds.com/ownership/howto/articles/45 792/article.html
A lot of people are making comments w/o knowing that facts, based on their experience with US cars. I'm sure I know less than the European drivers, but having just spent 3 weeks driving in Europe with a Renault would like to point out a few interesting facts.
..
My Renault (not the same model) used a keyless start system. The Renault representative told us to PUT the card key in a slot and press the start button. Good advice, but as we found out later, the key didn't have to be in the slot. When the key was anywhere near the car, the car could be started - no insertion was required. I would imagine a lot of drivers wouldn't know this - the Renault rep didn't - so if you pulled out the key in a panic and the car didn't stop, well .
My Renault didn't have a conventional parking brake. It was a fully automatic system that turned on when you turned the car off, and disengaged once the car was moving forward under power. To manually engage it you have to push a button on the dash and a LED would light up.. The Renault representative couldn't explain to us how it worked ("I never think about it - it's automatic"). It was not obvious to me if it could be used for an emergency stop.
The car would unlock whenever the car key was nearby. You couldn't tell if the car had been locked, because when your hand was placed on the door latch and the key was nearby, it unlocked. If you wanted to make sure the car had locked, you have to toss the key a few feet away, then goback and test the door.
I hard a hard time finding out how to turn the electrical system off - everything ran even with the key removed, and the engine turned off (lights, windshield washers, radio). Opening the door seemed to shut everything off. I couldn't find a manual way to do it. I also couldn't find how to turn the electrical system on w/o starting the engine - that is, there was no "accessory" position. I had to start the engine if I wanted to roll up a window I'd left open.
I couldn't read the French manual, but how many drivers read the manual anyway? This car just wasn't obvious to me.
Now, I had a 6-speed manual, but I wonder what an automatic transmission would be like in a car like this? Maybe some push button on the dash? Maybe neutral is automatically switched to under a certain speed?
In short, don't make the mistake of thinking that the subject car operates anything like an American car, and don't expect even a Renault rep to understand how it works. If the car has been designed to be fully automatic, the driver may not know, or may have forgotten, how to do certain things manually. Do you want to read the manual at 120 KMH?
I'd certainly agree with that sentiment - whilst I do think the captain must accept some of the blame [his preparation appears to have been poor and legally he is in command and responsible for the aircraft and its passengers], the question of the aircrafts response to his control inputs hangs in the air. Having said that, I still think the post-crash environment ensured that his failings were highlighted whilst any supposed flight-control system failings were not given their due prominence.
As for your comment about a nose down command, according to the captain he was requesting up elevator, presumably to gain height. I don't have the information about whether he had sufficient engine thrust at the time to gain height with an increased angle of attack - it is possible that the flight systems countermanded his command and held the elevator position until the engine thrust/airspeed components were sufficient to allow the elevators to move upwards without a stall occurring - in that instance its hard to fault the software, its just protecting the aircraft from an unsafe condition. In a manual system, the stick-shaker is to advise of an imminent stall, but it doesn't prevent it. [I can remember doing stall-training in a piper cherokee - even in something that docile its amazing how fast the stall comes after the controls start to get mushy].
I'm not an expert in flight-control systems, they no doubt make aircraft safer, more fuel-efficient etc [for passenger aircraft anyway], but that accident may show an instance when fly-by-wire systems 'cautioness' may actively prevent getting out of trouble - after all the line between diaster and escape is often very thin. The A320's system is an incredible piece of engineering, but perhaps it tended to provide a security blanket - I would imagine any pilot flying that low and slow in a non fly-by-wire would be sweating buckets [if it were possible, which I doubt].
p.s. Thanks for a great discussion - very involving !
It comes back to the "Pull the stick back to make the houses get smaller, push it forwads to make the houses get bigger. If you want the houses to get smaller then bigger really really quickly, keep pulling back"
Heavy airliners flying close to stall angle of attack (stall is conditional entirely on AoA only - you can stall in any attitude, at any airspeed just as soon as the critical AoA is exceeded) are very draggy. It doesn't matter what the engines are doing - at his AoA, increasing the AoA any more - by nose up elevator for instance - will quickly result in a stall. In something like a B727, at this point the stick pusher commands nose-down elevator by physically pushing the whole yoke forwards. In a sidestick-controlled FBW Airbus, the stick ain't going anywhere even if the elevators do start to command nose down to prevent the stall. (Crashing from even 30 feet stalled is far worse than crashing unstalled - the rate of descent is likely to be less if you crash unstalled).
A heavy airliner operating at a high AoA (i.e. very high drag - the Airbus possibly flying in a higher drag configuration than it ever would under normal flight conditions) will take a long time to accelerate and start climbing. The only thing the Captain could do in his situation is wait for more airspeed as IIRC he was pretty much nibbling critical AoA anyway, and any more nose-up attitude would actually decrease the amount of lift. But at 30 ft AGL with trees in front of you, it would be extremely difficult psychologically to do anything other than haul back harder on the stick, even if your training is screaming at you about stalling. (A friend of mine managed to stall the top wing of her Starduster Too at 50 feet AGL, and she said the most difficult thing she's ever done is actually lower the nose at 50ft AGL with the plane threatening to drop out of the sky. It must be said a 2-seat biplane that weighs under one tonne will recover and accelerate much faster than an A320).
Oolite: Elite-like game. For Mac, Linux and Windows