Sudden Unexpected Acceleration today

[Daniel in SD]

I had a crazy thought, hear me out:

All new cars automatically accelerate on their own at a linear rate and the ONLY pedal is the brake pedal. One. Big. Pedal.

Can’t have unintended acceleration unless of a total and complete failure of primary and backup braking systems! You modulate acceleration by using the brake pedal: Take the foot off the brake, accelerate like a P100D Raven. Slowly let the brake pedal come up, car starts to move at walking pace.

Done. Problem solved.

Uhmmn. Seems like having only an accelerator pedal would be much better. It could be the same size as the current pedal and Tesla could implement it in an OTA update. “True one pedal driving”

 

[Cirrus MS100D]

Bumper car mode! Brilliant!

 

[AlanSubie4Life]

Uhmmn. Seems like having only an accelerator pedal would be much better. It could be the same size as the current pedal and Tesla could implement it in an OTA update. “True one pedal driving”

Would take some getting used to. Would not want to release that pedal suddenly!

Good ideas all around here, though.

 

[Daniel in SD]

Would take some getting used to. Would not want to release that pedal suddenly!

Good ideas all around here, though.

It would require very good fine motor control. Seems like the pedal would have nearly twice the acceleration range (-1.2g -> 0.8g vs. -0.3g -> 0.8g). I would like them to implement this just to try it out. It would probably be great for autocross or track driving!

 

[AlanSubie4Life]

It would require very good fine motor control. Seems like the pedal would have nearly twice the acceleration range (-1.2g -> 0.8g vs. -0.3g -> 0.8g). I would like them to implement this just to try it out. It would probably be great for autocross or track driving!

It seems like a great idea; I'm surprised no one is doing it already.

 

[Oldschool496]

violent

I did not see any violence in the video, I saw a nice controlled stop to zero.

It takes about a second or two to engage the parking brake,

Thanks for the video, I always wondered.

"Warning" I am sure the brake lights don't activate, so even testing it, one should be aware those behind ever will not get a warning. That why they call it Emergency Braking.

 

[Az_Rael]

One starts out near 5V and one starts near 0. As the pedal is pressed, one goes up and and goes down

OK, I dont know beans about Teslas system which was why I was curious what the various folks here who have actually deep dived the logs thought.

So for the layman here, what would the 0% max pedal travel sections indicate in Figure 1? If the time stamps are accurate, that doesn't look like actual pedal stomps then releases then stomps. Unless the minimum time period reported in the data for those sensors is 1.0 seconds, so it's catching stomps and releases that were happening in less than 1 second and reporting them as 1.0?

 

[Oldschool496]

Just one type of surprise

Yes, one type of surprise of many. Very good.

 

[Oldschool496]

U

Always exciting as my P85 punched it while I was rounding a right turn.

UH, yes I get that quite a bit from loaners "Classic 85s" I'm using the cruise stalk as the directional, because its location differs then my refreshed S or later S's perhaps, as the TACC and Autopilot stalk are both in different areas now.

Its happened a couple times in cars I think with no Autopilot later added on many of those cars.

 

[Runt8]

For those who haven’t done it in your Tesla, FFS, go out and do it where it’s safe! It takes about a second or two to engage the parking brake, which slowly brings the vehicle to a stop. It doesn’t cause the car to skid out of control and it should obviously only be necessary in an emergency where the brake pedal isn’t working for some reason (e.g. your ex triggered a remote evaporation of your brake fluid)

Here’s an old example, but virtually nothing different with the current S/3/X

Doesn't the parking brake use the same brake pads as the normal brake? Older Teslas used to have a separate system but pretty sure that's been gone for awhile. So if your normal brakes aren't working I don't see how the parking brake would be any different.

Don’t forget the cruise stalk! I used to get unintended acceleration events in my classic S when I would engage the cruise stalk instead of the blinker. Always exciting as my P85 punched it while I was rounding a right turn.


Out of curiosity, what do you think of this fellow’s data? https://www.autosafety.org/wp-conte...udden-Acceleration-Log-Data-What-It-Shows.pdf

Real accelerator pedal sensor data from the log file of a Tesla S sudden acceleration incident has been provided. After giving an explanation for how the data was obtained from the two accelerator pedal position sensors, the data was examined. It was concluded that changes in the accelerator pedal sensor output caused the sudden acceleration. The alternating peaks and valleys in the accelerator pedal sensor data appeared to show that the driver was stabbing at the accelerator pedal. This led Tesla engineers to conclude that the driver was the cause of the sudden acceleration. But the four valleys indicating that the accelerator pedal was released had a duration of one exactly second accurate to 0.1 second. This accuracy in the duration of four successive pedal release times could not have been caused by a human driver. Instead, the peaks and valleys in the output of the accelerator pedal sensor appear to have been caused by a thermally induced leakage current in the motor speed sensor, which is housed inside the hot drive motor housing, and which shares a common+ 5V power and ground with the accelerator pedal position sensor. An explanation of how this leakage current originates, and how it increases the output of the accelerator pedal position sensor while the accelerator pedal is released by the driver, was provided. The explanation reveals that the four identical pedal release times are caused by the +5V regulator powering the accelerator pedal and motor speed sensors going into reset at high temperature, turning off the sensors for a fixed time duration of exactly one second. While the motor speed sensor is turned off, the algorithms controlling the vehicle’s drive motor become unstable, causing the motor speed to increase without control. This causes sudden acceleration to occur without the driver pressing on the accelerator pedal. This explanation provides a testable theory of sudden acceleration in the Tesla S vehicle. The theory applies to all Tesla S and Tesla X vehicles as a result of commonality in the design of their traction motor/inverter assemblies.

Ronald Belt is a bit of a quack. If his theory were in any way possible then we would be seeing all sorts of strange accelerator behavior as soon as the motors heated up. He's got a lot of assumptions but very little actual data and zero ability or desire to test his theories.

 

[Oldschool496]

So if your normal brakes aren't working I don't see how the parking brake would be any different.

Yes your right the older versions were two differing systems of pads and calipers: One regular brakes actuated by hydraulic pressure via the brake pedal and entirely different set of brake pads and caliper actuated electrically by the stalk.

Since I think its required to continue to have two entirely differing types of actuation for at least the rear brakes on vehicles, the fact now that its one set of pads and one caliper, must mean that one set is hydraulic for the pedal as well as electrically actuated for emergency braking.

The newer version is no different then the old days when the rear drums were actuated by Hydraulic pressure and just in case Hydraulics failed, there was a cable/tension actuation for emergency braking tied right to the drum system for Emergency braking procedures.

 

[Az_Rael]

Doesn't the parking brake use the same brake pads as the normal brake? Older Teslas used to have a separate system but pretty sure that's been gone for awhile. So if your normal brakes aren't working I don't see how the parking brake would be any different.

Ronald Belt is a bit of a quack. If his theory were in any way possible then we would be seeing all sorts of strange accelerator behavior as soon as the motors heated up. He's got a lot of assumptions but very little actual data and zero ability or desire to test his theories.

Ah, didnt know his history. I just thought the initial graph was interesting and wonder what would cause those 0% pedal travel in the data he shows. I suppose it could also be loss of log data for those time periods.

 

[Runt8]

Yes your right the older versions were two differing systems of pads and calipers: One regular brakes actuated by hydraulic pressure via the brake pedal and entirely different set of brake pads and caliper actuated electrically by the stalk.

Since I think its required to continue to have two entirely differing types of actuation for at least the rear brakes on vehicles, the fact now that its one set of pads and one caliper, must mean that one set is hydraulic for the pedal as well as electrically actuated for emergency braking.

The newer version is no different then the old days when the rear drums were actuated by Hydraulic pressure and just in case Hydraulics failed, there was a cable/tension actuation for emergency braking tied right to the drum system for Emergency braking procedures.

I didn't realize it was a requirement for a backup for actuation (although it makes sense). Thanks for the info!

 

[mongo]

Ah, didnt know his history. I just thought the initial graph was interesting and wonder what would cause those 0% pedal travel in the data he shows. I suppose it could also be loss of log data for those time periods.

That 0% is likely the output after the signal conditioning. So if the pedal sensors are out of range (like no power) it spits out 0% (the safest value) and also sets a DTC.
In other words, the number is not pedal travel, it is rationalized accelerator command.

 

[BigMod3Fan]

Here's a very good article on the subject: Opinion | Sudden Acceleration Often Caused By Drivers

I sent in a bug report and called Tesla right after my incident last Sunday. They asked me to call back on Monday when the team that looks at the logs can do a deeper dive. I did that but then they said to make a service appointment instead. So now, I have one set for Sep 10. Will let you know what they find out. To their credit, they are sending a service person to my home.

While it's entirely possible that I (and others) were victims of our own human errors, I am allowing for the possibility of a software glitch for the following reasons.

I have entered my garage hundreds of times without a single incident. I have entered my work parking space hundreds of times without incident. But under relative less common conditions such as making a tight turn into a parking space that has obstacles (fence or garbage cans in my instances), these incidents have occurred. Furthermore, I had obstacle aware acceleration and low speed collision avoidance on. As for the argument that the foot absolutely has to be on the accelerator for the car to accelerate that is obviously not true, otherwise summon would not work.

I will do my best to get to the bottom of this and if Tesla can prove that it's human error I will accept that and will share that with the group. In the meantime I would appreciate hearing from others who have had these experiences and how they approached Tesla about it. Hopefully, between all of us we can get to the root cause of the problem (human or otherwise) so that we can continue to help Tesla become THE iconic car company of the future.

 

[BigMod3Fan]

I could actually see driver error like this being worse with a Tesla, or EVs in general, due to regen and "one pedal driving". With aggressive regen people are less likely to need to press the brake when coming to a slow stop. This creates a bit of complacence when it comes to positioning your foot over the brake. Which means if you come in a bit hot and actually need the brake suddenly you're more likely to have your foot positioned over the accelerator creating the runaway panic condition that results in these accidents.

My question is why doesn't Tesla's automatic collision avoidance prevent this? It should know when you're actually about to hit something and simply stop you from doing so. I know this could create some weird issues with unusually tight parking spots, but make it a default option that people who run into those weird issues can disable if needed.

In my case I was 3 feet from the obstacles, going VERY slowly with no need to brake suddenly when the car lurched forward. At such a low speed the car should have stopped in front of the obstacles because of the low speed collision avoidance and obstacle aware acceleration settings.

 

[BigMod3Fan]

Was there ever a resolution to this post? I read page after page but didn't find out what the evidence showed.

Obstacle aware acceleration doesn't prevent you from accelerating. It just limits the output power. You can still run over garbage cans and you can still crash into a concrete wall. Just with a little less force.

Again, as has been proven to death.. There is so much redundancy and error checking in the pedals that unintended acceleration is really really really really unlikely to the point of being impossible. There have to be a lot of systems going belly-up at the same time, all reporting the same incorrect information. There is just no way.

Some have advice to turn on creep mode, as that would require you to press the brakes during parking.

I do use creep mode and the acceleration that I experienced was extreme and sudden, not something that I would expect if the output power was limited. I mean this thing absolutely bolted toward those garbage cans. Because I use creep, my foot was neither on the brake or the accelerator pedal when the car took off.

 

[Dan203]

In my case I was 3 feet from the obstacles, going VERY slowly with no need to brake suddenly when the car lurched forward. At such a low speed the car should have stopped in front of the obstacles because of the low speed collision avoidance and obstacle aware acceleration settings.

Where was your foot? On the brake? Or were you depending on regen to stop you?

I wonder what would happen if regen suddenly turned off? I know that under certain conditions the software will disable regen to protect the battery. If that happened and you were actively using regen to slow down it could feel like sudden acceleration as the resistance from the regen motors was removed.

 

[BigMod3Fan]

Where was your foot? On the brake? Or were you depending on regen to stop you?

I wonder what would happen if regen suddenly turned off? I know that under certain conditions the software will disable regen to protect the battery. If that happened and you were actively using regen to slow down it could feel like sudden acceleration as the resistance from the regen motors was removed.

At these speeds regen does not work so I can initially just coast into the parking space. So the acceleration happened while my foot was neither on the brake or the gas pedal. Like I said earlier, I have used this method to park in my garage hundreds of times and never been fooled by the transition from regen to no-regen. Believe me, I have played out all these scenarios trying to replicate the problem. I even tried to hit the gas pedal (in a controlled setting) instead of the brake but can't reproduce the sudden aggressive lurching.

 

[Tam]

...I do use creep mode...

With ICE and creep, it's effortlessly to get the car moving. Just take all your feet off and it creeps.

That's why I turned the creep off after I switched from ICE to Tesla.

Without creep, it actually takes effort to move the car. When I approach a parking spot, I would take all my feet off and the car slows down and I actually have to use the accelerator to move it to the correct spot. In my case, the "wrong" pedal would be the brake as it would stop the car from moving to a correct spot. Thus, it is really easy for me to know which pedal I am on, thanks to no creep mode.