Toymota

I'm bumping this thread because something similar happened to me!

But here's the kicker: it was in a totally different car! A rented Vauxhall Insignia (Buick Regal) in the UK. And it had a manual transmission. And it was a diesel.

This is my recollection of events:
I was coming to the end of a very long 170 mile, 3 hr highway drive from Cardiff, Wales to London Gatwick Airport. I had just exited the freeway, and was in the nearby second roundabout to turn into the airport. Just as I was taking my exit from the roundabout, I was shifting from 2nd to 3rd gear. But while the clutch was in, I realized that the engine was revving too high to make the shift. So I completely lifted my right foot from the accelerator pedal. It was not really possible for me to stare at my feet to make sure because I needed to look up and steer the car, but my right foot was lifted and not caught on the brake or accelerator pedals.There were NO floormats in this car.

However, the revving continued. For a split second, I contemplated whether I should put it in 3rd gear and release the clutch to put some load on the engine and maybe slow it down. But I decided it was safer to put it in neutral and release the clutch. So at this point, I was coasting at around 10 mph towards the exit. Both feet were off the pedals. But the engine was still revving. Keep in mind that this was a diesel so it only revs up to about 4,000 rpm.

There was a car behind me (luckily he didn't honk or anything) so I had to do some rapid troubleshooting. I wondered if the accelerator pedal was stuck so I started tapping then pushing it with my right foot. It was up, but then after a second the revving stops! So I put it in gear and continued driving.

It all happened over maybe 10 seconds and no harm was done. So I just returned the rental car and didn't mention anything about it, just in case they thought I was a crazy American who didn't know how to drive stick.

One other thing was the cruise control system was on but not set. I had used it on the highway but canceled it with the cancel button as I was nearing my exit. I had normal speed control at this time. This car also has an indication. The cruise icon lights up white when the system is on, and green when the speed is set. The icon was white before and after this incident.

So I dunno... what do you guys think?

chikoo

I entirely believe this. The more components in a system, the more points of failure arise.
Especially with a software, it is never ever completely predictable. Been working with electronically controlled machines and high end software long enough to arrive at this remark.

Kurtz

iTrader: (1)

Except electronic control means less components.to cause these issues.

Notice how nearly every genuine documented case of unintended acceleration (that isn't driver error or someone trying to score a payday) is caused by a physical'mechanical cause?

The floor mat in the case of the Lexus/Toyota incidents (and the wrong floormats, not secured properly, no less)

In older cars the carb or TB linkage could get caught on things.

Ford is currently having some early 2000s Tauruses investigated for runaway acceleration issues- caused by? A loose cruise control cable, which can get tangled with gas pedal cable

Toymota

There were no floor mats in this car.
It was a current model car so it probably had an electronic throttle.
I had both feet off the three pedals. I wasn't even trying to brake. So pedal misidentification was unlikely.

You sound like the type of person who sees everything in black and white and there must be a definite "physical'mechanical cause" to this, or a 'smoking gun'. I am actually willing to accept that sometimes we simply don't know what caused it, so we should leave the matter inconclusive and open for further investigation.

It seems like the car enters a certain operating mode on the highway, perhaps a lean burn mode, but when it comes off the highway it doesn't know and doesn't change to the correct mode. This could result in an incorrect throttle opening if it is trying to achieve a certain air/fuel ratio. I've also heard of it the other way, where the car comes off the highway and it stalls or idles too low because the throttle is closed incorrectly.

TooFly4Ya

Im thinking Cruise Control has something to do with these issues, regardless its not cool and needs to be fixed ASAP.

Kurtz

iTrader: (1)

I understand that. That's why I said nearly every case it was a physical/mechanical issue.

(note I was replying to someone else at the time too)

Not at all. There is always a cause. I'm not saying it's always physical, but I'm saying that in nearly all documented/investigated cases the cause was one of the following:

A) A physical interference issue. Sometimes floor mats, sometimes cables or linkage. In these cases switching to electronic controls would eliminate the problem.

B) Driver error (which is why reports of this happen so much more often in inexperienced drivers or old people)- in this case there's nothing actually wrong with the car to fix.

C) People lying. This was documented in a number of cases with people tyring to cash in after the big toyota media coverage. The guy with the runaway in CA who insisted he tried to stop and did all kinds of stuff, and then the computer records showed he was lying... the woman in upstate NY who insisted she slammed the brakes to stop, and the computer records showed she never touched any pedal but the gas... Again the car isn't at fault.


Please understand, I'm not saying YOUR case was one of those. I don't know what your case was. You chose not to report it to the owner or follow up though, so we probably never will know.

I'm saying that nearly all cases are one of those 3. And in the only one of those three where the car is at fault, electronic controls would have avoided the problem.

The NHTSA put in hundreds of hours investigating Toyota (and others) and found zero cases... ever... where electronics were to blame.

Until folks start showing up with genuine, documented, actually investigated to a conclusion, cases where the electronics are actually at fault, the continual tendancy of folks to wave their hands and go "ooh! these complex electronics! machines are gonna kill us!" is silly and groundless.




the car doesn't control the air/fuel ratio with the throttle. It controls it with the fuel injectors.

So the engine revving would have nothing to do with the car trying to maintain an A/F ratio.

At no point in your original description do you mention touching the brake though- what kind of sounds like happened is an issue with the cruise control...the brake would turn off the cruise generally, but otherwise it would keep the throttle revved to maintain speed.. Possibly when you starting tapping around the gas pedal in your troubleshooting you tapped the brake and that reset the system.

So maybe an error in the cruise system whereby you hit cancel and it turned off the light but then resumed the cruise speed (there's a resume button on most cruise systems- the least "car is possessed" explanation would be you hit that by accident, easy enough on a rental you're not familiar with, and the "set" light just didn't come back on... or maybe it doesn't come back on for "resume" only "set"...though that'd be a bit odd)


Again I'm not saying that's what happened, but it fits what you described.

Toymota

Hey no need to butt heads here. I'm open to discussion.

I did consider the possibility that I accidentally resumed the cruise control. It is a small rocker switch on the steering wheel.

[EDIT] It is on the left side of the steering wheel. My left hand was on the gear stick. My right hand was on the wheel and did not cross to the other side. It was highly unlikely that anything could have touched the resume switch. [/EDIT]

I had canceled the set speed several miles before approaching my exit from the freeway. I braked and downshifted to negotiate the offramp and first roundabout off the freeway. Everything was normal here. Do note that in a manual, pressing the clutch also cancels the set speed, which I did a couple of times but the engine was still revving.

I don't know that much about lean burn or air/fuel ratios. Only that I thought having a wide open throttle would be used to reduce pumping losses and a lean burn condition.

chikoo

Are you serious Klutz? oops I mean Kurtz ...j/k

But are you serious?
Less components? Have you ever seen a programmable electronic chip? How about many of them together to form a system? Have you ever looked at software code? Modules that are made to work fine on their own, do not play very well when used in combinations that the developers did not even dream of.

I guess not, otherwise you would not make such a comment.

chikoo

Obviously, you have NEVER experienced any complex electronics or software programming in your life.

lobuxracer

iTrader: (2)

OK, let's stop a couple of things right now. First, name calling. Second, I don't know who doesn't understand diesels, but there is no throttle plate in a diesel at all. Controlling a diesel is all about fuel control. Diesels take a full gulp of air on every cycle or they wouldn't work. They don't adjust AFR, they only inject enough fuel to produce the power you are asking the engine to produce, and the timing of the injection is critical to avoid detonation.

It's far MORE likely this car had a fuel control issue, but I guarantee you it wasn't related to a throttlebody problem because diesels don't have a throttlebody at all. They are not anything even remotely like a gasoline engine in terms of how they are controlled.

And yes, I have direct experience programming automated electronic test systems from way back in 1982, so I have a pretty clear picture of what can and does go wrong.

Toymota

Thank you lobuxracer for setting things straight. That is good information to know.

Kurtz

iTrader: (1)

And FWIW, I'm a software engineer for a living. Both development and testing.

So I'm pretty durn aware of the issues.


And if you have competent people writing and testing the stuff, it's far less likely to cause unintended acceleration than having a bunch of physical/mechanical parts that have to reliably work for tens and hundreds of thousands of miles and never physically get stuck or get in the way of something else.

And computers and electronic chips have been showing up in cars throttle and engine control systems for decades now, so it's not like this is a new thing someone just came up with.

Rough idles, weird shifting behavior, nav systems losing your location, crazy steering behavior? All documented issues related to computer control (though the steering one is less the computers fault and more the poorly shielded aftermarket items sitting nearby...)

But the car just randomly deciding to think the gas pedal is floored when it's not? Not so much.

There's a lot of failsafes to avoid exactly that from even being possible.

Just look up how many things the guy who faked the toyota acceleration story for ABC had to alter to make the car do it. He had to cut multiple different wires, then connect two other not-normally-connected wires, plus throw an additional not-found-in-the-car resistor into the circuit too. Which seems lot less likely to accidentally happen in the real world than, say, a physical cable or linkage getting caught on something.


As evidence I present, again, the fact that just about 100% of documented cases of actual unintended acceleration have a physical, not electronic, cause.



My apologies if continually injecting real world facts interferes with how confusing some of you think electronics and software are.

chikoo

My apologies Kurtz. I really did not mean to offend you personally. You have given a lot to this forum and I really appreciate that.

The key word you used in this reply is "competent". That is a human factor.
Over and above, when it comes time to test, you can only test what you know. It is over time that ones learns all the idiosyncrasies of a complicated system such as this. Even when the Toyota/Lexus support group analyses the issue, unless they have at least one dedicated person who joins heaven and earth in finding the root cause, all they will end up doing is test the car against a set of predetermined test scenarios. This is a catch 22 situation. You are testing the car against what you programmed it for. That will work in 99.99% of cases. In this case, obviously the car has experienced a new scenario, and failed to react properly to that. How can that be accounted for in standard run of the mill QA test?

Let me give you an example from the time when I was supervising a CNC machine shop (and new to the electronics programming world).
The item is question was a gigantic 3-axis CNC machine. For some reason, it had acquired a gremlin that it would start moving on one of it's axis randomly, spoiling a lot of jobs.
We had the machine manufacturer called in and they went over the machine like a tooth-comb over the electronics and related programming for over a week, unable to replicate the issue. Sounds all to familiar, right? They left with a QA success certificate. Next week, the machine started it's act again and this time, I ordered the machine operator to open up all the panel so that I can look at them. On a hunch, I told the machine operator to unplug the other end of the harness for that axis which was deep inside the machine, clean it and plug it back in. He did that and the machine never failed for the next 3 years while I worked there.

Not saying that this is the answer for the Lexus/Toyota gremlins, but what I am emphasizing is that you cannot program for everything when you develop the system. You make improvements in the system as you learn. Obviously this is one gremlin that needs to be figured out. Probably Lexus has figured it out, but for purposes of a mass panic,they may not release the information, but fix it as the cars come in for service, secretly, under routine service update.

Dble Agent

iTrader: (2)

It happened to me when i had my 07 IS350...but my flip flops were still holding down the gas pedal....this could be the same issue.