Back in the '80s it was Audi. Unintended acceleration caused by the fact that the transmission was connected to the throttle valve by way of the kick-down linkage. Every time, the person had their foot on the brake, but the car overpowered the brake and shot through the garage or across the street or whatever.
As soon as it made the news, people started coming out of the woodwork. Audi engineered a recall that consisted of a switch that prevented the transmission from being shifted from park or between forward and reverse unless the brake pedal was depressed. It did nothing to correct the "transmission problem."
Then people started reporting similar events with GM cars that used vacuum kick-downs, so it couldn't be the same problem. I was one of them. I knocked down my friend's parents' garage door, but when I was done, I looked down at my foot and found it firmly planted on the gas pedal. I had been pushing as hard as I could, because my mind was convinced that I was pushing the brake. A short time later, it was announced that this had been the case every time, although some people still swore that they had been pushing the brake. Now all cars have the brake/transmission interlock.
Funny thing. If you compare the torque that brakes can resist, to the torque that an engine can create, the brakes win every time. Anyone ever seen someone "power braking" a (rear wheel drive) muscle car? You hold the brake lightly with one foot, and nail the gas with the other. The front brakes have more power than the engine, so the car doesn't go anywhere, but if you get the it just right, the rear brakes still haven't engaged, and you can smoke the tires indefinitely because even partial application of the front brakes prevents the car from moving.
So how does a car that has a top speed of 90-100 mph without brakes achieve 90 mph when the driver is standing on the brakes?
Oh, and also, although the Prius has regenerative brakes, it also has mechanical brakes that are totally independent of the engine control system, so you have them even if the engine is out of control. But the gas pedal is fly-by-wire. THERE'S NOTHING CONNECTING IT TO ANY MECHANICAL LINKAGE! So when someone claims that it was physically stuck to the floor and they couldn't pull it up... Sorry. Can't happen.
So, while there may be something real going on here, maybe an issue with the cruise control. Have you ever set the cruise control at, say, 60 mph, but then used your foot to accelerate to 75, and then forgot you were in cruise control? I'm betting it's something like that.
Like I said. I'm waiting.
Posts: 3735 | Registered: Mar 2002
| IP: Logged |
posted
When I engaged the cruise control on my '94 Buick, the gas pedal would mechanically move as the computer adjusted the throttle and tried to control the speed. I could feel it under my foot, if I touched the pedal with my foot.
I'm pretty sure that my current car, a Toyota Matrix, doesn't have that feature while under cruise control. I'm going to have to test it to be sure, but if I remember correctly, the pedal returns to a neutral position and doesn't move while the cruise control is active.
Your points about the human factors in these accident reports are quite valid.
Posts: 1813 | Registered: Apr 2001
| IP: Logged |
posted
There was a story on NPR the other day that featured computers in cars as the main topic of debate. Basically, they were saying that computers are prone to failure, and that with more and more of our car's functions controlled by that computer, we're just asking for these types of problems, which are software related, not hardware.
I'd also note though that the guest on the program said that despite these problems, the safety and efficiency gains made from using computers fair outweighs the problems.
By the by, if your car starts to uncontrollably accelerate, why wouldn't you just put it in neutral or reverse? Wouldn't you roll to a stop when the momentum died?
Posts: 21898 | Registered: Nov 2004
| IP: Logged |
posted
If the same computer that controls the accelerator also controls the clutch, and it has a crash, it might not respond to an attempt to shift gears. Who knows how the Toyotas are built, though?
Posts: 1813 | Registered: Apr 2001
| IP: Logged |
quote:Originally posted by Lyrhawn: By the by, if your car starts to uncontrollably accelerate, why wouldn't you just put it in neutral or reverse? Wouldn't you roll to a stop when the momentum died?
I'm completely lost on this one too. Neutral's designed to be the most accessible gear on any transmission. How did a CHP officer go for whole minutes without once touching the transmission?
Regardless of Toyota's acceleration problems, there should never be a single fatality from this. If the car has the time to accelerate to a fatal speed, the driver has the time to drop it into neutral.
Posts: 3293 | Registered: Jul 2002
| IP: Logged |
quote:Funny thing. If you compare the torque that brakes can resist, to the torque that an engine can create, the brakes win every time. Anyone ever seen someone "power braking" a (rear wheel drive) muscle car? You hold the brake lightly with one foot, and nail the gas with the other. The front brakes have more power than the engine, so the car doesn't go anywhere, but if you get the it just right, the rear brakes still haven't engaged, and you can smoke the tires indefinitely because even partial application of the front brakes prevents the car from moving.
The dynamics are different when moving. At speed the brakes will quickly heat up and lose effectiveness. This is referred to as "fade." Most people don't see this much because they don't use brakes that heavily but I've experienced it during some less-than-responsible driving in a 300ZX and that car has pretty beefy brakes, specifically to minimize the effect.
A follow-on effect to the sort of heat that leads to fade can be brake pad glazing which can further effect stopping performance.
Whether fade is sufficient that brakes could lose out to a full-throttle input, I'm not sure. But power braking is a different beast.
quote:By the by, if your car starts to uncontrollably accelerate, why wouldn't you just put it in neutral or reverse? Wouldn't you roll to a stop when the momentum died?
I believe that automatic transmissions in many modern vehicles are mechanically decoupled from the gear shift lever. When you flip the lever to neutral you are merely requesting that action of the transmission/computer rather than directly inducing it.
Posts: 3275 | Registered: May 2007
| IP: Logged |
posted
You should still be able to put it in neutral, I think. One time I was coasting down a hill in neutral, and I meant to put it back in drive once I reached the bottom. I accidentally put it in reverse instead. I hit the gas, but the engine just revved and it stayed in neutral.
Posts: 9945 | Registered: Sep 2002
| IP: Logged |
quote:Originally posted by Lyrhawn: By the by, if your car starts to uncontrollably accelerate, why wouldn't you just put it in neutral or reverse? Wouldn't you roll to a stop when the momentum died?
I'm completely lost on this one too. Neutral's designed to be the most accessible gear on any transmission. How did a CHP officer go for whole minutes without once touching the transmission?
Regardless of Toyota's acceleration problems, there should never be a single fatality from this. If the car has the time to accelerate to a fatal speed, the driver has the time to drop it into neutral.
Regarding occupants of the vehicle, agreed, assuming the vehicle will shift to neutral. Pedestrians and others would be in slightly more danger though.
Posts: 369 | Registered: Apr 2007
| IP: Logged |
quote:The dynamics are different when moving. At speed the brakes will quickly heat up and lose effectiveness.
That is true, but the brakes are still more powerful than the engine, even at speed, until they begin to fade. In any case, the car was going 94 mph, which isn't possible with the driver applying any reasonable brake pressure.
But that's not the only part of the story that doesn't ring true. The accelerator pedal could not have been held to the floor, because there is no mechanical linkage to hold it to the floor. Even if the return spring broke, the only thing holding it down would be gravity.
This particular driver claimed "I pushed the gas pedal to pass a car and it did something kind of funny. ... It jumped and it just stuck there,"
In his 911 call he said "my accelerator is stuck. Yeah. I pulled it back, I tried pulling it back, pulling it back, but it's stuck"
He was told by the 911 operator to put the car in neutral, but he responded "I'm trying to control the car!"
Eventually, after slowing the car down to about 55 mph with the addition of using the emergency brake (which only engages one shoe of the rear drum brakes, and would have very little effect compared to the front discs) he put in in neutral and came to a stop with the cop car in front of him.
So in answer to the question about the transmission, yes, you can put the car in neutral, he just didn't do it.
Posts: 3735 | Registered: Mar 2002
| IP: Logged |
quote:By the by, if your car starts to uncontrollably accelerate, why wouldn't you just put it in neutral or reverse? Wouldn't you roll to a stop when the momentum died?
I believe that automatic transmissions in many modern vehicles are mechanically decoupled from the gear shift lever. When you flip the lever to neutral you are merely requesting that action of the transmission/computer rather than directly inducing it.
My car is a 2002 and whenever I accidentally push it into neutral, since I tend to rest my hand on the gear shift from time to time, it stops accelerating. Modern cars also automatically shift into neutral when you put the car in reverse and it's going too fast to compensate without causing problems.
Posts: 21898 | Registered: Nov 2004
| IP: Logged |
posted
According to Toyota, if the computer receives a signal from the accelerator pedal at the same time the brakes are being pushed, the computer cuts off power from the engine control system.
Posts: 3735 | Registered: Mar 2002
| IP: Logged |
posted
I flipped my Prius into neutral while accelerating a couple of times yesterday to test it out. Once with the accelerator almost floored. It worked fine.
Of course, it is an electronic function, not a direct mechanical one, so if that part of the computer goes wacky it might not work . . . but assuming it does, I don't see how it would make the car any harder to control.
Posts: 9866 | Registered: Apr 2002
| IP: Logged |
posted
I offer no opinion as to how valid her testimony is, but the woman I saw speaking about her Toyota accelerating out of control claims that she did try putting it in neutral and reverse with no effect. Maybe she was so freaked out that she imagined trying this.
posted
On the Prius, at least, you have to hold the control knob over in the neutral position for about a half second for it to work. If she just pushed it over and let go it wouldn't have done anything.
Posts: 9866 | Registered: Apr 2002
| IP: Logged |
quote:My car is a 2002 and whenever I accidentally push it into neutral, since I tend to rest my hand on the gear shift from time to time, it stops accelerating. Modern cars also automatically shift into neutral when you put the car in reverse and it's going too fast to compensate without causing problems.
Right, when it's all working correctly. I was suggesting that a computer malfunction might cause this to not work on a "shift by wire" setup.
Posts: 3275 | Registered: May 2007
| IP: Logged |
posted
What I can't figure out is why these people didn't just turn off the car. I don't care if the pedals stuck to the floor and breaks have failed and the transmission is locked in overdrive, can't you still just turn the key and shut off the engine?
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
quote:Originally posted by The Rabbit: What I can't figure out is why these people didn't just turn off the car. I don't care if the pedals stuck to the floor and breaks have failed and the transmission is locked in overdrive, can't you still just turn the key and shut off the engine?
That is something I doubt most drivers have ever done while the vehicle is in motion. I imagine in an emergency situation when your mind is evaluating all it's options, it simply ignores that possibility in favor of more familiar responses.
Posts: 14316 | Registered: Jul 2005
| IP: Logged |
posted
The Prius doesn't actually use a key. You shut it off by pushing the power button. It's possible that in a computer failure that wouldn't work. Also people are afraid that the steering wheel will lock if they shut thr car off. Toyota suggests holding the power button in for 5 seconds which puts it in accessory mode -- all functions except the engine remain active.
Posts: 9866 | Registered: Apr 2002
| IP: Logged |
quote:Originally posted by Glenn Arnold: Ok, I waited. Now it starts.
That article barely scratches the surface of what's fishy with that particular story. Check this out.
Posts: 9945 | Registered: Sep 2002
| IP: Logged |
quote:Originally posted by Glenn Arnold: Ok, I waited. Now it starts.
I saw that article. I know it must have been scary, but I really don't think that he could have been hitting the break on the car.
Posts: 1214 | Registered: Aug 2005
| IP: Logged |
posted
When so much of a car's operation is controlled by computer, what might be the effect of transitory EMPs induced from nearby power lines, or from all the communication signals being broadcast by drivers, and broadcast TV signals? Perhaps what is really needed to fix the reported problems is proper EMP-shielding, which basically means enclosing the computer circuitry in a Faraday cage, like the military does for its modern jet fighters.
Posts: 3742 | Registered: Dec 2001
| IP: Logged |
posted
Here is the story by Fumento that was quoted by one of the other articles above. I think one of his arguments that makes the most sense is, that Sikes claimed he was afraid to take a hand off the wheel to put the car in neutral (one of several reasons he claims he didn't do it when 911 repeatedly told him to) but he did apparently take his eyes completely off the road to reach down and pull up the accelerator, also removing at least one hand from the steering wheel. Posts: 1214 | Registered: Aug 2005
| IP: Logged |
quote:Originally posted by andi330: Here is the story by Fumento that was quoted by one of the other articles above. I think one of his arguments that makes the most sense is, that Sikes claimed he was afraid to take a hand off the wheel to put the car in neutral (one of several reasons he claims he didn't do it when 911 repeatedly told him to) but he did apparently take his eyes completely off the road to reach down and pull up the accelerator, also removing at least one hand from the steering wheel.
How did he manage to call 911 without taking a hand off the wheel? Did he have a voice activated hands free mobile phone? I suppose its possible but not that likely.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
posted
Depends on jurisdiction maybe, driving with one hand on a phone is illegal here. Consequently, sales of hand-free (often Bluetooth) headsets have apparently spiked.
Posts: 7593 | Registered: Sep 2006
| IP: Logged |
posted
Even with bluetooth headsets you still have to touch it to get it started, even when using voice activation. You have to touch a button on the headset to start the voice activation. Regardless, his story doesn't add up.
Posts: 1214 | Registered: Aug 2005
| IP: Logged |
But theoretically, you don't even have to take your hand off the wheel with certain Toyota models since the car itself uses Bluetooth and can pump the sound through your car speakers and the activation button is on the wheel. http://www.toyota.com/html/bluetooth/Posts: 7593 | Registered: Sep 2006
| IP: Logged |
posted
The thing that really bothers me about the whole "well you know, it's a computer so it might have screwed up idea" is that the computer in a car is essentially a sophisticated control system. As someone who works with controls I can tell you that one of the cardinal rules in control design is to always FAIL SAFE. Sure, you can miss something, but something as simple and straightforward as multiple redundancy in preventing a runaway vehicle has GOT to be high on the list for the designers.
So I have a hard time buying "the computer did it".
Posts: 409 | Registered: Apr 2002
| IP: Logged |
posted
All it takes is a random static charge--even a cosmic ray--flipping a bit here and there to produce a malfunction in a computer, so the original program does not perform as intended. Those unacquainted with computers seem to think they are incapable of error just because they are machines. But in fact, computers constantly malfunction and get altered data; this is why redundancy and checksums and sometimes even multiple computers with consensus comparison of parallel processes are employed so widely, and especially in critical applications in science and industry (such as spacecraft).
Posts: 3742 | Registered: Dec 2001
| IP: Logged |
quote: The car's front brakes showed significant wear and overheating, Toyota said. That kind of wear and heat would be consistent with the brakes being lightly applied over a long period of time, executives said at the press event.
Data from on-board computers indicated that Sikes had applied the brakes, to some degree, at least 250 times during the 23 mile incident, Toyota executives said, and that the brakes worked normally each time.
Edmunds.com has independently tested Prius cars similar to Sikes' and confirmed that the engine would stay engaged if the brakes were only pressed lightly, but not hard enough to actually stop or slow the car, said Dan Edmunds, head of auto testing for the automotive Web site Edmunds.com.
quote:According to Toyota, while Sikes' front brakes were worn away, his rear brakes were "fine," and a reading of electronic data from Sikes' car showed that he had applied the brakes and the accelerator alternately at least 250 times.(emphasis mine)
Ummmm...yeah. This guy is bankrupt, behind on his lease payments and the car's computer shows he was alternately accelerating and braking. I think that it's unlikely a computer malfunction would both cause the car to accelerate out of control and record that the car was alternately accelerating and breaking at the same time. I call foul.
Posts: 1214 | Registered: Aug 2005
| IP: Logged |
Printer-friendly view of this topic
![[Confused]](confused.gif)
