As someone who has been in software for over 30 years, I have to ask Toyota this question, “Have you ruled out the computer?” As I understand it, the cars in question are all “drive by wire” cars. Unlike the cars I drove as a teen where the gas pedal was connected to the carburetor with metal rods, these days the pedal is an input device that merely indicates to the car’s computer how fast the driver wishes to go. The computer is what directly controls the engine speed.
Even if the gas pedal was actually sticking, the computer could be programmed to detect a faulty pedal and shut down the engine automatically upon fault detection. Such software would be trivial to implement and could save lives until the real problem is determined. However there seems to be evidence of cases that defy the pedal theory and Toyota should be looking at the computer.
Computer code:
If (gas on floor) and (break on floor) then kill engine;
What do you think?
This seems very plausible to me. i wouldn’t be surprised if it turns out to be the case.
I don’t know the details about how the pedal is “sticking”. I had heard that Toyota had out-sourced manufacturing of the pedals to another company, and this company was having the problems.
That said, how exactly do you program a computer to detect a “stuck” pedal? If I have my foot on the pedal, at a steady pressure, will the computer think it’s stuck and shut the engine down?
#2 fairly simple, Brake pedal pressed while accelerating?
Yes->fault
I don’t think there’s a very easy way of detecting a bad gas pedal sensor. If the computer sees a certain voltage coming from the pedal, it’s going to supply the fuel and air needed. I’d say two sensors would be a good addition, but that’s lot harder than a easy software flash, not to mention cost.
The issue however is that since it is drive by wire one could change the software to detect that. If the computer thinks that the pedal has been floored to ling and senses the brake is applied the computer can shut down the car.
Seems to make sense. Just give it a certain voltage value. The only thing is the computer might not have a physical connection (or even wireless) connection to the brake sense. Then again, it might, if it works closely with the ABS module (unless it is the ABS module as well, which would be perfect, really)
My guess is that it’s a mechanical problem with the electronic throttle pedal causing it to stick intermittently. All modern electronic throttle systems will set faults and go into a reduced power limp-home mode if any discrepancy in the pedal position sensor and throttle position is detected. Any non-linear activity of the throttle position sensor or pedal position sensor should be flagged and cause limp-home to be activated.
I’m also a professional software engineer (tho I use the term loosely) and work in the field to integrate that software with the hardware (buildings, in my case).
I suspect this is a hardware issue, and unfortunately the software is not able to detect it.
Of course, one could install a simple “brake overrides accelerator” software decision. Flash the chips on the suspect vehicles and send people on their way?
The issue is with the carpets stopping the accelerator from springing back up when pressure is released. This problem could happen in most any car if the carpet is in the wrong spot for some reason or if the wrong carpet is used. Used to happen in our used BMW where someone had put in an aftermarket floor cover that couldn’t attach to the floor very well. Took me a second to realize what was going on and it was quite tricky to get the rug to stay where it belonged. EASY FIX: remove the part of the rug that obstructs the pedal path.
It’s always the damn software. 🙂
I learned this in high school: if the accelerator sticks, shift to neutral and turn off the damn car. How hard is that?
From what I have read this is how other manufacturers with drive by wire systems work. The computer will shut off the throttle if the computer detects that both pedals are being pressed. I also am starting to believe that there are 3 issues with the Toyota’s 1. carpet causing trouble. 2. The throttle assembly sticking. 3. Some engine computer issue.
http://www.caranddriver.com/features/09q4/how_to_deal_with_unintended_acceleration-tech_dept
#10. Uh…try telling that to my wife. First I would have to explain the word accelerator..OK Gas Pedal baby, and it goes downhill from there.
It’s not that she’s dumb, technically challenged, maybe.
I have a 2001 Lexus GS 300. After years of driving fords, I bought a Toyota.
The one of the first things I did was to run into the freezer in my garage. Problem, in dress shoes it is possible to hit the brake pedal and the gas pedal at the same time. They are much closer together than the ford’s pedal.
Issue two: My toyota (Lexus)uses an electric motor to pressurize to power brakes. The way it works is an electric motor spins up about every five pedal pushes. Pedal travel increases for each press until the motor spins up again.
I have found that if you drive a long time without pressing the pedal, the pedal travel increases as the pressure fades. Thus going into a garage or parking space and heel-toeing can result in a poor driver catching the side of the accelerator with the outside of their (insert: flip flop,croc,hooker heels,ugly flat front kicks)footware.
How many people were texting or just unskilled drivers?
Toyota is not the only vehicle with this problem.
My 03 Chevrolet did it. The mechanic said it was the idle control valve in the fuel injection. I have been told of BMWs (by a certified BMW mechanic) doing this for the same reason. The blame in these cases, from the guys that fix them, is cruddy gas coking the innards of the FI system.
If the fault is in the engine control module, does this software have to be validated in the same way that it is with software controlled medical devices?
Speculation is not needed (except if Toyota lied). It is called “stuck gas pedal”. Only possible interpretation is physically stuck gas pedal. Also, manufacturer of the pedal assembly is one investigated not only by Toyota but by Ford too.
Deeper problem is recent over-dependence on electronic systems in cars. There they’ll never be built at the quality levels of planes. Hence, whole load of problems will emerge from them (just imagine software/hardware glitch on radar equipped cars with “preventive” braking, slamming on the brakes at high speed on busy hwy for no real reason…). There should be NO automated “thinking” of the car allowed in operations that affect how is it driven. This includes killing engine when both gas and brake are pressed. This is not computer game but real World and real people. Such stupid electronic “preventions” are partly guilty of at least one “stuck gas pedal” deadly accident: car had push-button on/off. With “geniously” designed “feature” that you must keep it pressed for long time to turn off the engine. Experienced driver who just drove this model first time tried to turn it off in a reasonable manner: by single-pressing the button… Which didn’t do anything.
Have he had ordinary key – few people would still be alive.
Why not turn off the ignition swith? On most cars, this action will lock the stearing wheel.
Perhaps cars need a good old fashun “kill switch” like Brit bikes back in the day. Wonder if the Norma M20 has this prob – YIKES.
(see http://www.youtube.com/watch?v=4uTjq1s843k)
Many drive by wire cars already have logic to shutdown the throttle if a simultaneous press of the accelerator and gas pedal is detected. My BMW 328 has that, for example. I tested it the other day and it works as you would expect: I floored the throttle and then pressed the brake pedal hard, and the engine dropped back to idle even though the throttle was still floored. Toyotas and Lexuses do NOT have this logic and that may be a major part of the problem… I predict that a firmware upgrade to add this feature will be part of the Toyota solution.
Tom
Better yet – if your asshole bites a hole through the drivers seat, kill the engine.
I own a 2009 Camry. Can not see that front floor mat is an issue. We have both original set and a wet weather pair. Even if not hooked properly to carpet, gas pedal is no where near the mat.
Problem sounds more mechanial to me and effecting cars with greater than 36K on them from what I have read. Praying for a quick repair.
As someone who owns a Toyota Corolla 2009 I also believe as a former automotive technician that the computer could somehow be led to believe the throtle position is at max position. This would give maximum fuel to the engine. I experienced a strange occurance a few months ago when it would not start. It had rained and the engine cranked but no start. Also the starter kept cranking even in run position of key?? I had to turn key to off in order to stop starter. The next day the car started fine. The problem never was recreated.
My point is that the computer may have seen a trottle position that was not idle position which would have prevented the engine from starting. Could this be a similar cause for the run away cars? Who knows?
From today’s AP article on the subject…
“Toyota offered its most detailed description of the problem: Condensation can form in the mechanism that connects the foot pedal to the car’s engine, causing friction that prevents the pedal from smoothly springing back when the driver eases up.”
From the description, it seems likely that this could be caused by corrosion over time, resulting from the relatively infrequent occurrence of condensation in the linkage.
Damn..what happens when my empty beer bottle gets stuck under the gas pedal?????
The suggestion of killing the engine if the brake and gas pedal are simultaneously in use is exactly what volkswagens do. If you put your foot on the brake while having your other foot on the gas the engine is cut until you release the brake.
Simple logic: the brake pedal overrides the gas pedal. If the brake pedal is being pressed, then stop sending fuel to the motor.
#23. I often drive with both pedals, sideways, around corners, at 90 miles an hour. Loosing power at apex because of a computer program, would kill me!
#23. i might add, the new VW DSG gear box on a GTi once decided to change down to second gear (from third) on a loose metal road hilltop corner, sending me in it over a cliff with a front end totally locked up. So much for software.
#25: Unless you are in the habit of applying full throttle and maximum breaking simultaneously for more than 3 seconds, the software kill is not be an issue. Also, when it does kick in it is very gradual and ramped so there is no abrupt transition.
Tom
Terrible, terrible idea. Are you going to comfort the families who’s loved ones died because the sensors bugged out during a dangerous situation? Maybe you are ready to hand the wheel over to a fallible computer, but whatever “automatic kill switch” you want to add to my car that is not directly controlled by the driver will be removed and bypassed by me.
If my car begins to accelerate uncontrollably, then I’ll cut off the ignition. If that doesn’t work, I’ll toss the car in neutral. If you honestly cant figure out to do that, then it’s only a matter of time before some other inane situation will end up killing you. Survival of the fittest includes mental fitness as well. Who knows, maybe the only reason you exist on earth is to serve as an example to others.
Spot DEAD on!!! Next up Lexus recall – all models. See ya Toyota, it was nice knowin ya.
In 1980 I was driving a VW Rabbit Diesel that ran away. Stray oil from air cleaner compartment got into intake. Since motor was diesel, turning off key did nothing. Pressed brake hard, clutch started to slip (diesel has torque). Got up to 100 MPH, depressed clutch, motor red-lined and ran out of air cleaner oil.
That was a fun car. You could start in in gear at takeoff in one step. I also remember shifting into reverse at stop lights on snow packed roads to stop faster.