Shift-to-neutral to get out of trouble - not a sure thing?

You pose an excellent question. When I listened to Mrs. Smith's testimony my first question to her would have been, "how long did you hold it in neutral?" Not the right time or place to ask such a technical question, but essential because you have to hold it there a second for it to engage. My question for Mr. Lentz, still unanswered, is, "why, in the context of the Pedal Entrapment Recall, will only Avalon, Camry, and specific Lexus models get the brake override, but not GenII Prius?" While GenII Prius does have a brake override when both accel and brake pedals are simultaneously depressed at freeway speeds, Avalon, Camry, etc also have this. They are getting a back-up system. Why not Prius?

With respect to the SoCAl CHP crash, if you want to read the NHTSA rpt and see the photo of the floor mat fused to the accelerator pedal, suggest you go to www.safetyresearch.net Click on Toyota Sudden Unintended Acceleration and then click on resource documents. You'll find it at Santee Crash Rpt. As has been widely noted this was a loaner car, the driver's side floor mat in this vehicle was from a Lexus SUV, not made for the vehicle involved and it wasn't hooked. That's a dealer that should be litigated into oblivion. Also at Safety Research, you'll find Prof Gilbert's prelim findings; fascinating reading.

 

If you press the "P" button while driving, the car will engage neutral immediately.

 

The report they generated seemed excessively biased without substantial factual support, so out of curiosity I tried to find who funds this corporation, SRS, that generated the report. Surprise - it's about 5 different litigation firms currently planning on suing Toyota.

Toyota Critic Safety Research & Strategies Founder Admits Report Funding Came From Firms Suing Toyota : Auto News

 

Gol Dang It. OK, you have now found yourself hooked to a real, live electrical engineer. I design and build electronic circuit boards for real, live systems. I do analog; I do digital; I've done microcode; firmware; and mucked with bigger systems. I do field diagnostics. I've shepherded major systems through EMI (Electromagnetic Interference) and EMC (Electromagnetic Compatibility) testing. There's plenty of other professionals who do all this kind of stuff for a living, but, for the moment, you've got me.

First: I went and listened to her testimony last night on C-SPAN's web site. She comes across as very believable.

Second: She wasn't doing anything special and the blame Cruise Control light comes on and the car takes off. Both feet on the brake. The car fights her all the way, and she's topping 100 mph.

I presume this is not a Prius-type shifter. She put it in neutral. The car didn't stop. She put it in reverse. OK, this thing is fly-by-wire. I can vaguely believe that some software type thought it wouldn't be a good idea to drop the transmission, in pieces, all over the roadway. But it should have shifted into Neutral, but didn't. Think that all the Toyota fly-by-wire shifters need a two-second pause?

So, for some reason, she manages to get the car slowed down after six miles of fighting this thing. She stops it in the median. Her husband catches up with her. No funky floor mats found.

Now, the kicker. She's out of the Devil Car. There's a tow truck driver there. He wants the car in neutral. The husband gets in there, does so - hey, the car is supposed to be off! He's not supposed to be able to do that? And then THE ENGINE STARTS ITSELF. And they have an affidavit from the tow guy that it happened.

Gentlepeople, my nickel bet of the day is that one or more of the engine computers in that car was screwed. OK, let's talk about that prof. Toyota's been going around and around saying that there can't possibly be any fault in the throttle or the electronics or whatever. He takes a three-hour close look and FINDS A SINGLE POINT OF FAILURE in the throttle controller. From what he said, the two analog SAFETY-CRITICAL voltages go into the SAME BLINKING DEVICE in the throttle controller. You DON'T DO THAT. Period. Devices get ESD (Electrostatic Discharge) hits sometimes during manufacturing. And an ESD hit can, depending upon the device, kill the thing, leave it working, or (and this is the fun part) damage it so it fails, later. Never mind that: Standard silicon processing always has defects on the wafer due to dust, etc. Testing catches most of these defects, but not all. When SAFETY CRITICAL stuff is put together, one keeps the analog hardware and, for that matter, the first level digital hardware, extremely separate until a point can be reached where any flaw, short to ground, short to VCC, short to adjacent, can be safely detected by software. And, from what he said, the limits on the software let malfunctioning chips or shorts NOT BECOME DETECTED.

How often does stuff like this happen? Man, chip faults are rare - BUT THEY ARE ALWAYS THERE. Look up "bathtub curve" if you're interested.

But never mind that - a busted throttle controller is NOT going to start that car. What this smells of is a semi-hung controller somewhere. How can that happen? Let me count a couple of ways..

1. Modern microcontrollers/microprocessor designs (heck, forget modern - we're talking stuff designed 30 years ago!) contain supervisory circuits that monitor under and overvoltage conditions. One filters the heck out of the power going into boxes in systems like these, not to mention employing dc-dc converters and the like to give further isolation. But I've seen flaky converters and EMI filters that don't before now. One usually designs with a trip voltage range that's smaller than the operation range of the electronics on the board - that way, if the voltage goes too high or too low, the supervisor circuit trips before the electronics malfunction. However, these supervisory circuits can be pretty fast acting on surges that last such a short time that they wouldn't affect the electronics - so people put filter capacitors on the supervisory circuits so they won't trip before their time. But, get the right fast moving surge in there, and the circuit won't trip and the microprocessor gets screwed.

When stuff like that happens, it's not guaranteed that the watchdog timers will stop working. Software and hardware state machines can get into weird, flaky modes where things work, sort of. Sort of poorly, no joke. Turn the power all the way off and back on again - and it works. Does all this sound familiar?

Engineers spend more time than I care to think with specialized test equipment and kits of little parts tuning this stuff, not to mention hardware simulation tools, and test, test, test. Hope Toyota wasn't taking any short cuts.

2. Software, software, software. If I had a penny for every loose pointer I've had to chase down or state machine that had a strange, once in ten year condition, I'd be a millionaire. Did you guys know that there's something like ten errors in 100 lines of code before checks? Lint (or the equivalent) catches 90% of that; a compiler catches 90% of that; code walkthroughs by live bodies with eyeballs catch 90% of the remainder; functional testing catches more; system testing catches even more. However, when you've got MILLIONS OF LINES OF CODE there's NO QUESTION that there's bugs in the code. The point being is that software is always shipped with bugs. The issue: Are the bugs non-service affecting? People spend their lives trying to get this stuff right, and, no, I'm not a CS major, although I've written my share of code by now. However, (and this is the important bit), there are a heck of a lot of supervisors out there with schedules in hand who are convinced that if they shorten schedules and put pressure on programmers to "ship it anyway", they'll get their bonuses at the end of the year. And, if those supervisors do it long enough, they can infect an organization with the same attitude.. Wanna lay bets about strange software? I don't know if Toyota is like that or not. But, one way or the other, we're going to be finding out.

There was more in that hearing. Did ya hear the one where the safety expert, analyzing NHTSA's own data, found out that there were multiple cases where cars with racing engines WERE AT THE DEALERSHIPS AND OBSERVED TO BE DOING THAT!?! Wth?


So, whatever it is that's going on, it's rare. That fits. But, when it does happen, Bad Things Happen. That can't be explained by a simple busted throttle controller. Well, I've got a Gen III Prius, and I'm not going to stop driving it, but, boy, I'm going to stock up on the popcorn when the criminal investigation starts putting papers out in court.

As far as the lady: Badmouth her and her husband at your peril. This is not the old-fogy-in-an-Audi-hitting-the-gas-instead-of-the-brake-and-going-fifty-feet-into-a-restaurant trick. Six miles? Give me a break.

Kbeck.

 

I intentionaly held the throttle down this am at or above 70 mph and also held start button down. The motor died at once and the trannsnission went to neutral. The power steering went off of course but at that speed steering to the off ramp was no problem. Was that so hard to do?, easier than turning the key off in my Town car. Whats the big deal about stopping a runaway car.

 

Right now I gotta take what these "victims" are saying with a pound of salt! Once lawyers and the possibility of mega settlements is added to the mix, stories and facts change!

 

The answer is: The Start Button is not connected to a magic box that puts a knife through the power feeds. It's connected to a computer. And that computer is likely connected to other computers. The only sure-fire way to stop a computer is to depower it. And what do you want to bet that most of the computers in the Prius never have their power disconnected?

I don't know. But once you start mucking with software systems, it's not a physical key disconnecting the power. It can be made safe - but that takes serious, hard work.

Basic point: If I took my old 1971 VW Beetle and turned off the ignition, the Car Was Off. No power to the components, She's Dead, Jim. Put a software controller on the On/Off button, get fancy: You have no guarantees beyond the say-so of the company that put the whole business together. You want that shifter to actually work? Software, thank you. You want the gas pedal to do something? More software.

At least the brakes and the steering on the Prius have a direct, physical connection to the appropriate hardware so, in the event of extremely dead software, Things Work Anyway.

The problem: That Lexus of that lady's. Hey, it turned itself on, by itself! Oh, yeah - the gas pedal seems to have a single point of failure! And Toyota was stonewalling for years. So, where's the trust?

Personally, I don't think my Prius is going to explode or anything. Maybe. And there's plenty of other cars out there running with fly-by-wire, and I bet some of them have worse problems.

Look at it this way: It's the feeling one gets when one's virginity goes that-a-way. One might be pregnant, or not, but it's not going to be known until the tests come back. The good news: If this was a real solid fault there'd be dead bodies all over or, more likely, the cars in question would never have seen the light of day. The bad news: Rare intermittents are hell to find and isolate. When, in my job, I find one of those pernicious things, and I've got the failed-then-but-it's-working-now boards in my hands, I go berserk with every repeatable test I can think of, and ask people to come up with ideas that I haven't thought of. I've had boards and components sliced and diced and looked at with electron beam microscopes in attempts to find out What Died, and why.

So, what was Toyota doing with roughly 2000 out of control cars? By the reports at that hearing, they weren't digging into the guts of those cars with soldering irons flying.

Dammed ostriches. People died because they were playing ostrich. Like I said before: I've got my popcorn handy.

Kbeck.

 

I've moved from a GS300 to a Prius. I had 'the mat problem' twice with this Lexus. I wasn't using Lexus floor mats but typical after-market auto store mats. These mats had a tendency to creep forwards, possibly because of the direction of pile, or maybe because of typical 'foot moves' - I've no idea.

I had no idea they could possibly interfere with the accelerator pedal. Never seen that on any car in years of driving.

The effect is not funny.

My first reaction was: why's the accelerator stuck? I scrabbled around with my hand around the accelerator, while still trying to control the car with the brakes. When I pulled the mat back, the accelerator was released. This didn't take very long, so I was lucky to find the cause so quickly.

It's all very well pontificating about what "I might do", but everyone's reactions to sudden and unexpected 'accidents' is different, and no-one can predict what accidents might happen, be it hoods suddenly rising, brakes failing, doors dropping off etc. OK I've never had that one. *

I think the suggestion to practice engaging in neutral, switching off etc in a moving car (safely!) is very practical.

* I once had a car with rear transverse leaf spring suspension, and bodywork bolted onto 'outriggers' attached to the central chassis. To stop the rear wheels from moving forwards or backwards, tie rods were connected to the outriggers. However, when one of these punched through its outrigger, due to rust and constant flexing, that particular rear wheel would move forward on braking, and backwards on acceleration: thus changing the direction of the car each time. Very alarming. Until I got it to a welder, changing gear was accompanied by heavy sawing motions on the steering wheel.....

 

If you put the brakes on firmly using both feet or just one the car will stop, period. It may take about 10% to 20% longer if the throttle is held wide open, but it will still stop in less than 200 feet from 60 MPH on wet or dry pavement. If you don't believe it try it.

Here is one good reference: http://www.caranddriver.com/features/09q4/how_to_deal_with_unintended_acceleration-tech_dept (Originally cited by Robbyr2 in another thread.)

Part of the C&D editorial describing tests of a V6 Camry, "With the Camry’s throttle pinned while going 70 mph, the brakes easily overcame all 268 horsepower straining against them and stopped the car in 190 feet—that’s a foot shorter than the performance of a Ford Taurus without any gas-pedal problems and just 16 feet longer than with the Camry’s throttle closed. From 100 mph, the stopping-distance differential was 88 feet—noticeable to be sure, but the car still slowed enthusiastically enough to impart a feeling of confidence."

 

I heard that woman's testimony. How can you be going six miles with the engine running at full speed without topping the car out at 110+ mph. By some miracle the car only runs at a certain speed?!? Give me a freaking break, these people are out there trying to make a buck or trying to hide their stupidity. The braking system are separate from the ECU, they are not related to the engine at all.

The cop was just as stupid, how hard is it to put the car in netural. He doesn't have enough time to think, yet enough time to call for help. Darwin....

I have raced against cops in a legal environment, most of them don't know much about cars.

 

Why do so many people declare the cop to be stupid because a passenger, not the cop-driver, called 911?

As for him not trying to shift to neutral, the last I heard that it was still an open question. A very few others have alleged that their traditional automatics couldn't shift to neutral, and I'm not hearing any adequate followup to this issue.

As a teenager driver, I had dad's older car jam in 3rd gear. Not a runaway situation, but I did question why he had never told me that that car had done so several times previously.

 

Hi All,

I read that Ms. Smith was driving a Lexus ES-350. This is not a hybrid. And from the Lexus website appears to have a plain-jane (but with gating guide) automatic transmission shifter.

I imagine to put it in neutral, one needs to press the button on the shifter knob and push forward, and to the left. The gate then would avoid allowing the shifter to go into reverse, as its positioned in the neutral position.

This DOES NOT appear to be a by-wire shifter. Am I wrong ? Or is this a by-wire shifter that is just made to look like a traditional mechanical shifter?

 

You don't need to hit the button to get into netural! to get a es350 up to 120mph plus, you have time to read a book and think about what to do

 

The front seat passenger dialed 911.