BMW 335i Unintended Acceleration?

This is a repost moved from the new member section. I would appreciate more comments from other Bimmers. This is not to bash BMW. I loved the car every minute I drove it. Drivers usually don't floor the pedal or the brake in this situation. Thanks in advance.
___________________________

I had a weird experience I never thought real last week. I live on a hill side in Southern California, and the road down to the town is a bumpy, narrow, and zig-zag with sharp U turns. When I tried to make one turn driving 335i with a foot covering the brake, the car suddenly boomed and accelerated. It stopped with the belly on the curb and wheels in the air after hitting the guard rail about 12-feet away across the street. I could have rolled over the hill if not for the guard rail, or run over people if it happened in a busy intersection. Everybody, BMW and insurance, seems to blame me as a reckless driver.

I have never got into accidents that I caused for over 30 years of driving. My another car is Porsche Cayman. I enjoy driving, but I feel haunted and scared of driving now. I experience nightmares every morning since before I wake up. I have enjoyed driving BMW 335i immensely for over three years. It is a small car with an explosive power. Nothing to find faults with. I cannot replicate the same situation, but I was not in panic and I am not too old not to tell the pedal from the brake.

Is there anyone who can shed some light on this happening? When I looked through Internet, there are lots of sudden acceleration cases of other brands, but not specifically on 2007 335i. It is the first twin-turbo model, and they have extended warranty for 10 years on the defective high-pressure fuel pump on this model.

You likely hit the wrong pedal.

While I am not sure about this, BMW likely uses the same (BOSCH?) drive-by-wire (DBW) throttle control unit on most of its engines. Therefore, an unintended acceleration issue with the 335i will also happen across the lineup. In fact, that same DBW unit is likely used by other automakers as well.

Not the answer you like to hear, but likely the truth.

- Mike

Originally Posted by wit3356

I tried to make one turn driving 335i with a foot covering the brake, the car suddenly boomed and accelerated.

Not likely. BMWs have a "smart throttle" that disengages when you apply the brakes.

You had your foot on the gas.

Originally Posted by KentW

Not likely. BMWs have a "smart throttle" that disengages when you apply the brakes.

Is that also true for a manual? If so, doesn't that mean heel-toe downshift doesn't work?

Maybe OP's shoe is too wide?

Thanks for your comment. Yes, I hope it should be my fault. Otherwise, it would be too scary to drive further.

Is any one out there who knows about the Drive-by-Wire? In my understanding, what they mean by wire here is the electrical wire. So all modern cars are made this way using more electronics. It may not necessarily mean that it is more reliable. We are relying on the electronic pedal position sensor and engine control module plus fuel injector, and many more. If they don't transmit or receive correct electrical signals, they may not function properly. This is the different case, but Toyota recalled recently 4 million cars including Lexus they sold to US since 2004. They are blaming the floor mat, but the case is pending. The recall period coincides with the period when more electronics began to be used in their cars. I am not making any case here, nor want only answers I would like to hear. There might be some issues related to the new Drive-by-Wire system.

Originally Posted by wit3356

Thanks for your comment. Yes, I hope it should be my fault. Otherwise, it would be too scary to drive further.

Is any one out there who knows about the Drive-by-Wire? In my understanding, what they mean by wire here is the electrical wire. So all modern cars are made this way using more electronics. It may not necessarily mean that it is more reliable. We are relying on the electronic pedal position sensor and engine control module plus fuel injector, and many more. If they don't transmit or receive correct electrical signals, they may not function properly. This is the different case, but Toyota recalled recently 4 million cars including Lexus they sold to US since 2004. They are blaming the floor mat, but the case is pending. The recall period coincides with the period when more electronics began to be used in their cars. I am not making any case here, nor want only answers I would like to hear. There might be some issues related to the new Drive-by-Wire system.

BMW introduced its first DBW system in the 1988 750iL, I think. The 3 Series got it in 1999, with the E46.

BMW's throttle safety systems are countless (and good), while Toyota's has been one cluster-fu*k after another.

- Mike

Originally Posted by TofuTurkey

Is that also true for a manual?

I will try it tonight (my wheels are already dirty).

- Mike

Thanks for your opinion.

Lexus and Toyota are respected brands to most people, and let's not downgrade them.

I am just curious whether throttle controls can be fail-safe or fault-tolerant. Computer systems used in airplanes, banks, hospitals, governments, ... are fault-tolerant, but it appears that consumer products are not. For being fault-tolerant, computers need 3 or 4 times more of the computing power, which I doubt is true in the consumer products. Although computers are more reliable than 20 years ago, but all PCs freeze as you know, which is a faulty condition. I am wondering how fail-safe the car electronics parts are. Any one out there to comment?

I recently attended a school on the "drive by wire" throttle control systems on all cars sold in the US. As you can imagine the main concern on everyone's mind, right after how we can make money on these things, was reliability and liability.
There are so many redundant (dual or triple) systems at both the signal and receiver end that runaway acceleration is theoretically impossible. At least two signals are generated at the pedal, one high voltage, one low voltage. These are carried by separate wires and the voltage decoded at the throttle. If the signal is whacky the system goes limp home. The throttle is also heavily safeguarded. It has a stepper motor which generates a signal every time it moves. A minicomputer compares that signal, coming from two separate sources within the throttle, with the pedal signal. Any mismatches? Limp home.
Even if all these signal matching safeguards fail, the throttle motor itself is set to fail in a closed position. Limp home. No, the motor does not have to close the throttle! If it doesn't, a spring does. Limp home.
I cut my teeth on this kind of scenario back in the runaway Audi days (simply put, drivers were hitting the accelerator and not the brake, but that almost put Audi out of business). I'm not naive, and have some ability to imagine how things can go bad. I cannot think of any possible way the current dbw systems can make a car take off when the throttle isn't pushed.

Edit: I believe that the dbw system has as many feedback loops and redundant systems at the hydraulic controls on an Airbus.

To be fair to AUDI... "60 Minutes" almost put them out of business with their original "news" story. I had an Audi 5000 at the time and the resale value of the car went to almost zero overnight.

Thanks for the great comment.

It is great to know that many fail-safe redundancies are built-in. The old Audi case is a different one. Is this Toyota case just another scam? The victim was a CHP (California Highway Patrol) policeman. The runaway car was Lexus ES350. His brother in the car called the CHP and shouted "Where is the brake?" before the fatal crash. His wife and a daughter was also in the car. If anyone knows better to stop the runaway car, the victim was the one, who got the professional training on that. So, the runaway is only to Toyota. Toyota blames the floor mat.

Before we go any further...was this car sold as a "prime loaner", i.e. essentially a new car. There are an 'increasing' number of reports of these cars spontaneously adjusting their suspension settings (camber and or spring heights) independent of the driver and shredding tires. This might be the same phenomenon in the throttle system?
.


[Emission: we need a "Troll" smiley in our repertoire. Something big and hairy]

Originally Posted by DSXMachina

I recently attended a school on the "drive by wire" throttle control systems on all cars sold in the US. As you can imagine the main concern on everyone's mind, right after how we can make money on these things, was reliability and liability.
There are so many redundant (dual or triple) systems at both the signal and receiver end that runaway acceleration is theoretically impossible. At least two signals are generated at the pedal, one high voltage, one low voltage. These are carried by separate wires and the voltage decoded at the throttle. If the signal is whacky the system goes limp home. The throttle is also heavily safeguarded. It has a stepper motor which generates a signal every time it moves. A minicomputer compares that signal, coming from two separate sources within the throttle, with the pedal signal. Any mismatches? Limp home.
Even if all these signal matching safeguards fail, the throttle motor itself is set to fail in a closed position. Limp home. No, the motor does not have to close the throttle! If it doesn't, a spring does. Limp home.
I cut my teeth on this kind of scenario back in the runaway Audi days (simply put, drivers were hitting the accelerator and not the brake, but that almost put Audi out of business). I'm not naive, and have some ability to imagine how things can go bad. I cannot think of any possible way the current dbw systems can make a car take off when the throttle isn't pushed.

Edit: I believe that the dbw system has as many feedback loops and redundant systems at the hydraulic controls on an Airbus.

+1 Well stated. People don't realize that it is possible to design technology with clearly defined failure modes. Traffic signals would be the best example that I always use because they 100% fail to a predetermined state...not all green.

Originally Posted by DavidLavin

To be fair to AUDI... "60 Minutes" almost put them out of business with their original "news" story. I had an Audi 5000 at the time and the resale value of the car went to almost zero overnight.

I did too!

And, my mom stepped on the wrong pedal and went through our garage door!

- Mike

Originally Posted by wit3356

Thanks for the great comment.

It is great to know that many fail-safe redundancies are built-in. The old Audi case is a different one. Is this Toyota case just another scam? The victim was a CHP (California Highway Patrol) policeman. The runaway car was Lexus ES350. His brother in the car called the CHP and shouted "Where is the brake?" before the fatal crash. His wife and a daughter was also in the car. If anyone knows better to stop the runaway car, the victim was the one, who got the professional training on that. So, the runaway is only to Toyota. Toyota blames the floor mat.

That driver fu*ked up. It is sad, but he simply made poor decisions at a critical moment.

Even if the brakes overheated and failed (likely), he could have put the car into neutral or shut off the engine. Most of us would still be alive.

- Mike

Originally Posted by DSXMachina

I recently attended a school on the "drive by wire" throttle control systems on all cars sold in the US. As you can imagine the main concern on everyone's mind, right after how we can make money on these things, was reliability and liability.
There are so many redundant (dual or triple) systems at both the signal and receiver end that runaway acceleration is theoretically impossible. At least two signals are generated at the pedal, one high voltage, one low voltage. These are carried by separate wires and the voltage decoded at the throttle. If the signal is whacky the system goes limp home. The throttle is also heavily safeguarded. It has a stepper motor which generates a signal every time it moves. A minicomputer compares that signal, coming from two separate sources within the throttle, with the pedal signal. Any mismatches? Limp home.
Even if all these signal matching safeguards fail, the throttle motor itself is set to fail in a closed position. Limp home. No, the motor does not have to close the throttle! If it doesn't, a spring does. Limp home.
I cut my teeth on this kind of scenario back in the runaway Audi days (simply put, drivers were hitting the accelerator and not the brake, but that almost put Audi out of business). I'm not naive, and have some ability to imagine how things can go bad. I cannot think of any possible way the current dbw systems can make a car take off when the throttle isn't pushed.

Edit: I believe that the dbw system has as many feedback loops and redundant systems at the hydraulic controls on an Airbus.

Great post! Nicely done.

- Mike

Originally Posted by xrated335

Before we go any further...was this car sold as a "prime loaner", i.e. essentially a new car. There are an 'increasing' number of reports of these cars spontaneously adjusting their suspension settings (camber and or spring heights) independent of the driver and shredding tires. This might be the same phenomenon in the throttle system?
.


[Emission: we need a "Troll" smiley in our repertoire. Something big and hairy]

Please forward head shot, asap.

Originally Posted by DSXMachina

Please forward head shot, asap.

The shot was there...

...I chose not to take it.

Here is the best I could get at short notice...taken with iCamera in my apple not 5 minutes ago

Hey xr, Happy Belated Guy Fawkes Day!

"Remember, remember the fifth of November,
The gunpowder treason and plot,
I know of no reason
Why the gunpowder treason
Should ever be forgot."


Taught to me by me mum

Originally Posted by Emission

That driver fu*ked up. It is sad, but he simply made poor decisions at a critical moment.

Even if the brakes overheated and failed (likely), he could have put the car into neutral or shut off the engine. Most of us would still be alive.

- Mike

According to the story, the car was rushing at 120mph with a flame underside. The victim probably did not know that the push start button should be held pressed for longer than 3 seconds to turn off the engine. Maybe the gear shifting didn't work. In 3 seconds, the car might have moved more than 500 feet. Even an off-duty police officer can not drive at such speed in San Diego highways.

Originally Posted by xrated335

Before we go any further...was this car sold as a "prime loaner", i.e. essentially a new car. There are an 'increasing' number of reports of these cars spontaneously adjusting their suspension settings (camber and or spring heights) independent of the driver and shredding tires. This might be the same phenomenon in the throttle system?
.


[Emission: we need a "Troll" smiley in our repertoire. Something big and hairy]

My 2007 BMW 335i was bought new with a few miles on it. It was a flawless car. I have driven it for three years with a little more than 20k miles on it. I didn't even have the HPFP problem.

I read that the runaway Lexus 350 was a loaner car given to the victim while his car was in service.

Originally Posted by wit3356

According to the story, the car was rushing at 120mph with a flame underside. The victim probably did not know that the push start button should be held pressed for longer than 3 seconds to turn off the engine. Maybe the gear shifting didn't work. In 3 seconds, the car might have moved more than 500 feet. Even an off-duty police officer can not drive at such speed in San Diego highways.

I feel sorry for him and his family. But, I really feel it was avoidable.

I just took my 335i 6MT out for a test drive. In 3rd gear at 60 mph I floored the throttle... then I pounded the brake at the same time with my left foot. The car immediately slowed to a hard stop and then stalled.

I couldn't tell if the engine was cut back or not, but it didn't fight the brakes. Funny thing, my Check Engine light and my Low Coolant light both came on (the car was pissed). I shut down and restarted the car and everything was normal again.

- Mike

Originally Posted by Emission

I feel sorry for him and his family. But, I really feel it was avoidable.

I just took my 335i 6MT out for a test drive. In 3rd gear at 60 mph I floored the throttle... then I pounded the brake at the same time with my left foot. The car immediately slowed to a hard stop and then stalled.

I couldn't tell if the engine was cut back or not, but it didn't fight the brakes. Funny thing, my Check Engine light and my Low Coolant light both came on (the car was pissed). I shut down and restarted the car and everything was normal again.

- Mike

Yes, it was a sad story. I live in San Diego, and it was a big news. Even Toyota Chairman apologized in a press interview saying something like that customers buy Toyota for its safety, but he had a remorse that Toyota lost the confidence of the customers. They indirectly admitted that modern electronic cars can run away.

I agree that brake will overide the gas during the normal operating condition. But it appears that the accident results from intermittent faults. I experienced a few times the Check Engine Light when starting. If I tried again, usually the car started ok without any trace. So here is the catch. If the fault is not fatal, people forget about that. If fatal, they are no longer around to witness that. Even though they survive the accident, the court may side with the manufacturer since the fault cannot be replicated.

Please don't try such dangerous stunt with the floored gas and brake. You know my Porsche has the Sport II option with a launch control. It is exactly what you tried. They can cut 0 to 60 time by 0.2sec. The launch control works like this. At the stop, you press the gas and brake together to rev the engine at the full throttle. When the green light is on, you just release the brake to launch the rocket. It is a dangerous option I haven't even tried.

Originally Posted by wit3356

Please don't try such dangerous stunt with the floored gas and brake. You know my Porsche has the Sport II option with a launch control. It is exactly what you tried. They can cut 0 to 60 time by 0.2sec. The launch control works like this. At the stop, you press the gas and brake together to rev the engine at the full throttle. When the green light is on, you just release the brake to launch the rocket. It is a dangerous option I haven't even tried.

Agreed. Most should not try this.

However, I do this sort of thing for a living. I'm an automotive journalist and I test (um... flog) cars for a paycheck. (I also race Porsches for fun.)

I've done dozens of "launch control" starts in the Porsche 911 Turbo PDK, Porsche 911 C4S PDK, Porsche Boxster S PDK, and Porsche Panamera Turbo PDK... and in many other cars. As a result, I am well aware of how vehicles will handle. In fact, the more I test... the more experienced I become.

In all honesty, I suggest you try your PDK "launch control" on an empty road soon. You will understand your car's capabilities, plus you will learn how to react under heavy acceleration. Testing is a good thing, and it prepares you for an emergency.

The Highway Patrol officer died because he did not completely understand the vehicle he was driving. He was unfamiliar with the controls and operation of the Lexus loaner.

Did we ever welcome you to Bimmerfest?

- Mike

Originally Posted by Emission

I did too!

And, my mom stepped on the wrong pedal and went through our garage door!

- Mike

I too about put a 5000 through our shop door one morning. I also installed what seemed like thousands of the gear shift interlock kits on the 5000's.

The issue not spoken to with these things was the large bellhousing which pushed the gas and brake pedals farther left than most other cars. This put the gas pedal damn near where the brake pedal is on most 'normal' cars.

The 5000's weren't/aren't the only cars with pedals farther left than expected. They just got the most attention...

To the OP, look at your gas pedal stopper and see if it is crushed in anyway. Audi never found a 'hurt' brake pedal system, but sure found enough crushed gas pedal stops. A clearer sign of cause couldn't be dreamt of...

Brian