Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Unintended Acceleration: Toyota ECU code somewhat short of perfect 2

Status
Not open for further replies.
"Not loose floor mats, not a sticky pedal, not driver error."

but...not proven /not/ to be the above.

I think they've shown the software isn't robust against all errors, but that's a big step from saying that as a result it has actually caused any of these incidents.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Nonetheless, the report would seem to indicate that a software error is at least as likely as dozens of people failing to notice that they were intentionally accelerating.

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529
 
Makes me think I'm not a good candidate for driving a Toyota.

My Daily Driver for-several-years Volvo 240 automatic's pedals are at a nice height, spacing, and position relative to the seat so that every now and then I inadvertently press both when going from "accelerating"(8D) to braking by pivoting my foot. Their response curves are such that amazingly the willing but wimpy engine power far exceeds the braking power for the first portion of the curve, so the sensation of RUN AWAY upon an application of brake is quite attention getting. My reaction to reposition my right foot has become somewhat automatic, but it certainly adds a bunch of braking distance.

I think I must have developed laziness or slightly deteriorated leg twisting motion.
I have long resisted left foot braking to avoid possible brain fade problems when changing vehicles, but LFB seems worthy of a closer look these days.
 
Sound like more content free conclusions... Can't remember if it was Car and Driver or Road & Track Magazine did a road test of ordinary to some of the most powerful engine equipped cars.. and with the braking systems on all of the cars with today's far better brakes, they were able to stop the car with the brakes with full throttle on the engine.... You would have to get on the brakes hard and bring it down.. Too much light wimpy braking could possibly allow the brakes to get to fade conditions..

I guess the days of expecting a semi-mechanicly literate driver to simply reach down and turn off the ignition switch are gone. Even as a young, inexperienced driver 16 year old kid, I had enough sense to kill the ignition while driving a 389 tri-power (3 carb'd 389) Pontiac with the infamous GM broken motor mount, which the owner had not told me about, when I picked up the vehicle. With the left side broken motor mount, it allowed the engine to lift up from engine torque and the throttle linkage was a perfect design for that motion to pull the throttle wide open.

I guess that would be asking too much of today's drivers... let alone throwing the transmission in neutral. The Start/stop button systems I familiar with do have hold down override to kill the engine..

Software errors, corrupt memory are not new problems, neither are the independent hardware techniques in real time controllers (independent from the main processor core e.g. dead man timers) to kill wayward code.


 
Doesn't turning off the key also lock the steering wheel on many cars?


Mike Halloran
Pembroke Pines, FL, USA
 
A more detailed report in EE times, I think it was, identified rather a lot of undesirable features in their code, including 11000 global variables and a failure to adhere to their own internal coding guidelines. It turns out the rather overloaded main routine, known as task X, was directly responsible for running the cruise control, which certainly has the authority to mess with throttle behavior.

Yes, a typical column lock engages about 15-30 degrees off centre, it could get ugly but in practice it seems OK.

Basically, if you have your wits about you, there is usually a solution, but if you have brain fade or panic then a stuck throttle can kill you.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
DanEE said:
I guess the days of expecting a semi-mechanicly literate driver to simply reach down and turn off the ignition switch are gone.
Click to expand...
When there is no key to turn, what do you do? When you're panicking because someone's rear bumper is approaching more quickly every millisecond, and you've already wasted several precious seconds slowly coming to the understanding that you may no longer have full control of your vehicle, how many here think they'll have the presence of mind to hold a button down for three seconds to kill the engine? That's a whole lot of ifs...

Dan - Owner
Footwell%20Animation%20Tiny.gif
 
Stuck throttle management should be a part of Drivers Ed Training...

All the cars I have had do not lock the steering column when the key is turned to the off position, but do it when the key is pulled out of the ignition switch.
 
Again, assuming there's a key to turn... I haven't had a key to insert into the ignition in quite a few years now. turning a key is a fast action... holding in a button continuously for 3 seconds seems like an eternity when you're barreling down a road about to crash.

Dan - Owner
Footwell%20Animation%20Tiny.gif
 
Yes there no primary safety advantage in a push to start push to stop button. In terms of secondary safety not having your right kneecap impaled on the ignition lock is a bit of a plus.

However these systems go in as a sales feature, not safety.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376
 
Never had an ignition lock that was anywhere near my knee but I have short legs. Maybe the safety krauts could outlaw the push to start buttons though I doubt it would reduce the number of unintended acceleration claims. Even with less than perfect code, it's a proven fact that most of these cases were human error, some "intentional error".

I had a 79 Mustang that the choke mechanism fell off the carb and jammed the throttle about 1/2 open. I managed to drive it the last couple of miles home turning the key on and off to modulate the power. Had to run a traffic light and parked it in the front yard. Fortunately it was past midnight and very little traffic.

I remember the car mag article mentioned above. Not only could every car easily stop from 60 with the engine at full throttle but the stopping distances were only increased by ~10 feet from the best possible stopping distance. They tested a wide range of cars, some quite powerful, the results were all very similar.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.
 
Some of the new features that cars have:
[ul] [li] Steering by wire [/li]
[li] Throttle by wire [/li]
[li] Brake by wire [/li]
[li] Wireless ignition enable (no key, just a START button) [/li]
[/ul]

And, under development are self-driving cars!

In a few years, "defensive driving" will mean you carry a gun and know where to shoot the vehicle to stop it.
 
Interesting to hear that the code has some defects. Any real-life testing proof that the code can cause SUA?

I though all the Toyota's in question are push to start but I believe they all still have a shifter you can whack into neutral.

I also believed that the US regulations on the anti-theft requirements were changed in the early 2000's to only require one anti-theft lock-out instead of two. My new truck doesn't have a column lock, just the shifter lock.

I was bored one day and read through a report entitled "Toyota Sudden Unintended Acceleration" released by Safety Research & Strategies, Inc.

The NHTSA complaint text for whole bunch of reported SUA incidences was included. The typical incidence text read along the lines of "I was driving in a parking lot and right went I pressed on the brake pedal to stop the car accelerated." My opinion of these incidences was that the people caught the gas pedal with the right side of their foot when they pressed on the brake.

If you don't think stepping on both pedals would catch you off guard then I suggest you try it in a vehicle where the gas and brake pedal are close together and at similar heights. You will likely find the engine easily over-powers the brakes when you step down on the brake pedal enough to get the amount of braking you have learned you need for the stopping condition at hand. Then, when you realize you're not stopping quite quick enough you will try to brake a little harder but the engine applies more power which defeats the extra braking effort.

With an automatic and the typical factory stall converter the engine only revs maybe 500-1000rpm before it stalls and the exhaust and sound damping makes the engine rather quiet so it's not like you'll hear this high rpm screaming engine to make you immediately realize your foot is depressing the throttle.

 
But the shifter is electronic, and would be controlled by the same processor that's potentially lost in code, or locked up.

In military systems, certain mission critical controls are not allowed to be passed through ANY software, or have at least one hardware-only redundant path. I haven't seen anything here that positively says that the power button has such a provision. If it doesn't, then turning off the ignition might not even be possible in a SUA.

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529
 
Any automatic I have driven still had a cable going to the transmission to engage the parking lock. Electronic shifted between gears but park, neutral, drive and reverse were mechanically selected.
 
One other thing to consider. The company that did the investigation was the expert witness for the first of many plaintiffs for lawsuits that could total 100's of millions if not billions. It's nor really in their best interest to find anything good about the Toyota firmware.

 
Status
Not open for further replies.

Similar threads

HungryDinosaur_imported
  • Locked
  • Question
Load Calculation of Dual Batteries
Replies
8
Views
1K
HungryDinosaur_imported
HungryDinosaur_imported
MechDevEng
  • Locked
  • Question
Sealing of an overmolded electrical connector pin
Replies
4
Views
1K
waross
waross
rQuestionEngineering
  • Locked
  • Question
The trickiest part of HVAC is just addressing the leaks at hidden locations.
Replies
1
Views
784
MadisonCameron
MadisonCameron
Cuocomponents
  • Locked
  • Question
a device that realizes the function of an ideal diode/OR-ing controller
Replies
3
Views
750
3DDave
3DDave
MRSSPOCK
  • Locked
  • Question
TWELVE DIODES INSTEAD OF SIX
Replies
12
Views
2K
bradrs
bradrs

Part and Inventory Search

Sponsor

Back
Top Bottom