Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

3 Unintended Acceleration events when parking my Tesla S P in my garage.

gajafl

New Member
Apr 24, 2020
4
6
punta gorda fl
I have recently had 3 unintended acceleration events with my 2020 Tesla S Performance in the past two months. Both events happened when parking my Tesla S in my garage. There was no driver error involved in any of these 3 incidents. We enjoy fast cars and our last four garaged cars were a GTR, C6 ZR1, C6 Z06, and C5 Z06. We keep near new car outside which is now a 2018 Pacifica with 4200 miles on it. Our two high performance garaged cars are use only once a week and we have done this for the past 20 years. I have a raced cars at the drag strip and I am also an Instrument Rated Commercial pilot. I mention all of this only because readers immediately insinuate that all unintended acceleration event claims are the fault of older drivers or those not familiar with fast cars. Two events occurred on August 18 at 990 miles and one on October 12 at 1250 miles. All 3 events occurred after our weekly Sunday one hour plus spirited driving ride in our Tesla. I typically use over 25% battery power for the one hour plus ride. Using Ludicrous Plus mode in FL 95 degree heat and making numerous 1/8 mile or longer runs causes both drive motors and the transmission to get very hot which in turn transfers this heat to attached sensors. We then take 20 minute slow ride in our local 20 to 30 mph zones to allow the transmission and engine coolant temps drop. This helped when using our GTR and our previous SC ZR1 before going home. I have since learned this slow ride in our Tesla S actually increases and not decreases the drive motor and transmission temperatures in our Tesla SP.

At 4:30 pm on August 18, the day of the first two incidents it was 95 degrees and the Tesla cooling fans were running when I pulled into my driveway and for 15 minutes after in the garage. We have two large floor fans which we direct at the car after being used. When in the drivewayI put the car in Park and disconnect my seatbelt which makes it easier to put my data Logger and radar detector away. I always reset the Driving mode to ‘Chill’ before parking my Tesla anywhere. I park my Tesla on the left side and my GTR on the right side of my garage. With my wife directing me I position my Tesla right front tire against the 3 inch high Riser on our 20” X 14” x 3/8 rubber Parking Mats. We have used these mats for 35 years to park our cars in our garages. With my Tesla right front tire against the Riser, my wife gives me a “Full Fist” sign. I release pressure on the accelerator which engages the brakes. At this location my Tesla front bumper is 19 inches from my workbench. I then push the Park button which activates the Easy Entry/Exit, and I exit the car. We have two large floor fans that we aimed at the nose of our cars to cool them after a ride. On this day the instant I pressed the Park button the car lurched almost 3 feet forward up and over the Parking Mats 3 inch Riser and hit my workbench. I instinctively hit the brake to stop the forward movement but the lurch happened in one second. My Tesla bumper had penetrated the workbench cabinets by over a foot. I looked up to see the shocked look on my wife’s face who stands just inches from the right front fender. The abrupt forward movement was so forceful and quick tossed the right Parking Mat backward four feet. I then backed up about 6 feet to exit the car but I was not in location where I was able to open my door due to my large tool chests against the wall. I then pulled forward to where I would normally park if using the mat, released pressure on the accelerator, pressed the Park button, and again my Tesla bolted forward 3 feet forward into the workbench. Twice in 5 minutes, the instant I pressed the Park button which automatically engages the brake, my Tesla had lurched 3 feet forward. Shocked again backed up about 10 feet and squeezed myself out of the car to inspect the damage. Hitting the workbench lightly damaged my Tesla front bumper because I had removed my License Plate Bracket. The incident did extensive damage to my workbench cabinets. We took photos of the damage to the car and workbench. We then wiped down and covered the car like we always do. The next morning I tried to duplicate the incident a half dozen times to see if I had a defective Parking sensor but I knew that the high heat from the ride was an integral factor in causing this event. I contacted Sarasota Tesla Service, reported the incident and sent t them a three page letter of explanation along with photos of the damage to the car and workbench. I have good carpentry skills but it still took me a week to repair the damage in the particle board laminate workbench cabinets and upper drawer assembly. Two weeks later we went to Tesla Sarasota and I had them inspect the damage to the car and inspect numerous other flaws I had reported after delivery. Sarasota Tesla forwarded my letter and photos to Tesla engineers who downloaded the EDR data from my car. I knew this would be a dangerous ongoing problem and would happen again without repair by Tesla. I told the Sarasota Tesla mechanic that this would happen again. He said Tesla engineers will contact me in few weeks. I had no idea how to prevent it from happening again and neither did the Tesla mechanic I spoke to. I have received a thank you call from Tesla for ordering a 2021 Plaid but no calls about my 3 unintended acceleration incidents.

We then used our Tesla 4 more times since the August 18 incident without any problem parking it in the garage. For these rides I left my seatbelt connected until after I pressed the Park button on the Parking Mat. I felt that disconnecting my seatbelt which activates Easy Entry/Exit is a contributing factor in causing this problem. October 10 at about 4:20 pm with the outside temp at 90 degrees we had a similar third unintended acceleration incident when parking our Tesla in the garage. On this day I had disconnected my seatbelt in the driveway. My wife asked me if it would happen again and I said it wouldn’t because I didn’t believe the outside temp was lower and the car was hot enough. I was wrong. Again I put my data logger and radar detector away. I had just recently sold my GTR because it was no longer as much fun to drive. I put its Parking Mat under the left front wheel of the Tesla. My wife helped me positioning both tires against the Parking Mats 3 inch Risers and this time I applied the brake before pushing the Park Button. Even with my foot on the brake, the instant I pushed the Park Button the car forcefully vaulted forward up and over the both Parking Mat Risers. The forward lurch was so violent it lifted the entire nose of the car up and over the both mat Risers. I applied maximum pressure to the brake the instant the car moved foot but it had no effect and the car bolted forward but this time stopped just one inch from my workbench. I couldn’t believe that it was able to climb over the two 3 inch Parking Mat Risers with my foot on the brake. I believe having to go over the two parking mat Risers absorbed some of the forward energy and had prevented the bumper from penetrating the workbench again by 3 feet. I was pretty sure but still praying I didn’t hit the workbench and was relieved when my wife told me I did not hit it. . It was apparent to me at this point that the car was hesitating going into Park, it remained in Drive for hundredths of a second with the brakes applied, and somehow there was an extreme momentary impulse to the accelerator that made the car overpower the brakes and move forward before actually going into Park. I then just sat in the car with my bumper just one inch from my workbench and waited for the cooling fans to cool the motors. I had my wife get my cell phone to take photos of the car while I sat in the car. I then backed up 10 feet and exited the car. The photos showed the left mat had been tossed back 5 feet and left long black marks on my garage floor. The right side Parking Mat had been wedged under the right front wheel. We finished with the photos, repositioned the mats, parked car, and wiped it down and covered it. I called Sarasota Tesla Service the next day and informed them of our third unintended acceleration incident. I sent them anther long letter and more photos. Again Tesla engineers have since download the data from the EDR and have not responded. The Tesla Support site states that Tesla is obligated to provide customers with a readout of any downloaded problems.

I know Tesla will never admit there is any problem with my car or its sensors. I don’t care about the cost of repair to my car or workbench and would pay Tesla to repair my car. I now better understand what the Tesla model S and possibly model 3 owners feel like that have filed 127 complaints about their Tesla having unintended acceleration events, some after the car was parked and they were walking away.

The Tesla computer does not generate a fault code when a problem like this occurs like all other cars with standard ECU’s. I have not yet heard from Tesla engineers about the first two incidents or now a third and will be surprised if and when they do. I know as we all do that Tesla will never admit that any unintended acceleration is possible in any Tesla and will never pay for any damage to a Tesla or property. I have had TPS accelerator sensor and PCM module problems in brand new cars in the past. In the standard ECU unless a fault code is generated by a defective sensor and makes the Engine Light stay on the mechanics have no way of finding the problem. I had a brand new 2008 Trailblazer SS with 1000 miles on it that quit 3 times from 5 to 15 minutes over a period of one year. It wasn’t until the last event that threw fault code and turned on the Engine Light were the GM mechanics able to repair the car. After a week at the dealership both TPS sensors had to be replaced. The PCM module causing a similar problem is story for another time.

I know that Tesla will never give me or the NHTSA the actual EDR downloaded data from my car or from any Tesla car. Rather the Tesla engineers will give just their interpretation of what the data says. Tesla will never admit that they have any unintended acceleration problems in any Tesla and certainly not recall any cars or pay for any repair to the car or personal property.

The NHTSA as of January 2020 states there have been 127 claims,110 crashes, and 52 injuries caused by unintended acceleration in Tesla vehicles. NHTSA wants Tesla to recall a massive number of cars and it will not. My own in-depth research into this type of unintended acceleration event in Tesla’s when parking is not new and has been clearly explained by Robert Belt, Electrical Engineer in a 2018 article. He was able to obtain thenEDR downloaded data from a Tesla model S that had a similar unintended acceleration event when parking he car. For those of you with a Tesla model S or model 3 I suggest you read Robert Belt’s report on these sites, (https://www.autosafety.org/wp-conte...udden-Acceleration-Log-Data-What-It-Shows.pdf) . (Tesla Owners Accuse Their Cars of Unintended Acceleration. Is It True? | Digital Trends) His articles clearly describe how the Tesla Transmission Sensor delays when overheated by the overheated drive motors from prolonged periods of fast and then slow driving. Tesla engineers are obviously aware of this problem. According to Belt this unintended acceleration is caused by an overheated Transmission Sensor which causes a delay in the transmission going into Park. The Tesla Transmission Sensor shares a common ground with the two (TPS) Throttle Position Sensors for the Accelerator. Due to the extreme increased transmission temperature caused by spirited and then prolonged slow driving increases the transmission heat and Transmission Sensor temperature it hesitates to put the transmission in Park by hundredths of a second. While the Transmission Sensor in trying to send Park impulses it sends as many as 4 voltage spikes to the TPS Accelerator sensors in one second. The downloaded data from the EDR from drivers who have this unintended acceleration problem indicates the driver hit the accelerator 4 times in one second which is impossible Tesla should certainly be able to fix this problem even without a major recall. Tesla claims that the software in the new Tesla S is 4 times more powerful than the software in the Tesla model S that was using autopilot when it ran into the side a white trailer truck jackknifed across RT 75 in FL in 2016. Authorities believe the driver was watching a DVD. No computer or ECU is infallible no one should ever trust one’s life to one.

The Tesla S Performance with its 825hp/967 ft lb tq using Ludicrous Plus mode and LC makes for a very fast stock car. Its 2.3 seconds 0 to 60 time, 6.5 second 1/8 mile time, and 10.4 in the quarter mile are phenomenal for any car let alone a 4 door sedan. The Tesla SP is the fastest new stock car I have owned. The build quality in terms of body gaps, panel alignment, paint flaws, cracked mirror housing, and surface flaws are disappointing but the performance makes up for these flaws. Tesla has amazing technology but any auto ECU’s or computer can only correctly process the information from the numerous sensors if given correct and timely input.

I doubt many Tesla model S owners drive their cars I do. I would be interested in hearing from any model S owners who have had any similar sensor hesitation problems or any unintended acceleration events. In spite of my ongoing issues with my Tesla S Performance I will still buy a 2021 Tesla S Plaid if it is the 9 second car ¼ mile car as advertised. I am hoping that the Plaid three motor configuration will eliminate the sensors overheat issues and prevent similar problems. I will enjoy my Tesla until I get my Plaid and make some changes in my garage parking procedure. The 2020 Tesla S Performance with the Raven updated battery is still a fun car to drive. Thanks.
 

MP3Mike

Well-Known Member
Feb 1, 2016
15,375
33,362
Oregon
NHTSA wants Tesla to recall a massive number of cars and it will not.

Saying things like this discredits everything else that you say.

  1. NHTSA has not asked Tesla to recall cars for SUA events.
  2. NHTSA has the power to force a recall, it isn't optional when NHTSA wants a recall.
NHTSA has opened a defect petition into the SUA events with Teslas, at the request of outside people. They have yet to grant or deny it. If they grant it they will open an official investigation.

As far as Robert Belt's theories, he has published two different "testable" theories, but hasn't actually tested them to see if they are valid theories or not.

According to Belt this unintended acceleration is caused by an overheated Transmission Sensor which causes a delay in the transmission going into Park.

  1. There is no transmission in a Tesla. (Just reduction gears.)
  2. As such, the transmission can't go into park. Going into park is just the two rear brake calipers being electrically acuated.
 
Last edited:
  • Like
Reactions: Silicon Desert

JohnSnowNW

Active Member
Feb 13, 2015
2,637
2,762
Minnesota
You didn't mention whether or not you had "Object-Aware Acceleration" on or not.

That said, I don't believe activating the parking brake caused your vehicle to lurch forward, despite that being how you remember the chain of events. That just doesn't seem remotely plausible.
 

MP3Mike

Well-Known Member
Feb 1, 2016
15,375
33,362
Oregon
My wife helped me positioning both tires against the Parking Mats 3 inch Risers and this time I applied the brake before pushing the Park Button. Even with my foot on the brake, the instant I pushed the Park Button the car forcefully vaulted forward up and over the both Parking Mat Risers. The forward lurch was so violent it lifted the entire nose of the car up and over the both mat Risers. I applied maximum pressure to the brake the instant the car moved foot but it had no effect and the car bolted forward but this time stopped just one inch from my workbench. I couldn’t believe that it was able to climb over the two 3 inch Parking Mat Risers with my foot on the brake.

You should really try this in an open, safe, environment: put your left foot on the brake and then floor the accelerator with your right foot and see if you can get it to shoot forward.
 
  • Like
Reactions: MorrisonHiker

MorrisonHiker

S 100D 2021.4.15
Mar 8, 2015
9,576
8,785
Colorado
You should really try this in an open, safe, environment: put your left foot on the brake and then floor the accelerator with your right foot and see if you can get it to shoot forward.
I assume he'll get the same error message that everyone else does when they test this safety feature, but it shouldn't hurt to try.
 

mtndrew1

Active Member
May 12, 2015
1,329
3,721
Gardena, CA
Teslas don’t have conventional transmissions or parking pawls as in a gasoline car.

Engaging Park activates a very strong brake on the rear wheels. You should be able to hear it when you push the park button.

As such it seems there’s a fundamental misunderstanding here about how the car works and way, way too much granularity about your other cars and garage and so forth.

If this is real and not pedal misapplication there are a few courses of action better than a verbose post here.

1. File at safercar.gov

2. Press the voice activation button and say “bug report” when something odd happens with the car. This will take a snapshot of vehicle parameters and a time stamp for future diagnosis.

3. Call the customer support line from within the car (tap the number on the car’s info screen) and ask them to remote in to the car and look for errors.

Edit: Which service center is the car with and what did the logs say?
 

wk057

Senior Tinkerer
Feb 23, 2014
5,659
11,430
Hickory, NC, USA
While I wanted to give a fellow pilot a benefit of the doubt on this, unfortunately, I had to mostly stop reading here:

I applied maximum pressure to the brake the instant the car moved foot but it had no effect

As noted numerous times in threads all over this forum in elsewhere, the motors are not powerful enough to overcome the brakes. If you've got the brake fully pressed, even if somehow the motors were trying to output full power you wouldn't go anywhere. I've tested this personally on several occasions, and it's impossible for even the most powerful version of Tesla vehicle available to overpower the friction brakes. This is by design. There's always instances where you need to stop more quickly than you could accelerate. Anyone can use the stopping distance numbers and such to calculate out the values and show that the stopping power far exceeds the vehicle's potential output power, and that's not even counting losses to the road and the like in those calculations.

I skimmed the remainder of the post...

Mainly, keep in mind that the Robert Belt paper is complete garbage and has no basis in reality. As I've noted previously, It doesn't even get close to correctly describing how Tesla's accelerator pedal sensing works at all. The setup described in that paper is something a hobbyist grade motor controller might use and has nothing to do with how Tesla's systems read this data. Without redoing my entire previous analysis (you'll have to hunt the forum, I don't have time at the moment), I'll briefly note that Tesla's system uses independent and fully isolated power supplies for the dual pedal sensors. These have a specific operating range. A value of 5V or 0V (as claimed in the paper) is just rejected outright, because this is outside the range of acceptable values for either sensor. Both sensors have to match exactly for there to be torque applied.

I know several people who have driven my Tesla vehicles and had brain farted over foot position at one point or another, thinking they were on the brake because the vehicle was slowing to a stop while they used only the accelerator pedal. There's definitely a subconscious disconnect with some folks, even in experienced drivers, over this type of control that takes some getting used to. If the car is decelerating, something in the brain sometimes tends to assume and believe your foot must be on the brake when it isn't.

In any case, I'll gladly extend my previous bet to the OP. We can schedule bringing the car and $10k in cash to my shop where we'll examine the vehicle logs together, step by step. If they show plausible user commanded acceleration, I keep your $10k and post the log analysis publicly. If they show no pedal press but power flowed and the car accelerated without user command, I'll double your $10k, post a detailed analysis that backs your position, and offer to testify to the same in whatever legal proceedings you pursue in the matter.

For the record and full disclosure, I'm undefeated on this. I'm a softy, and have thus far let everyone off the hook as far as the public shaming goes (as all have plead with me not to do this), along with not even beating them up over the cash aspect (even though I doubt I could expect the same from them had they "won"). At this point it's a waste of time for me, so I don't think I'm going to go easy on the next folks who lose this bet. I've personally examined quite a few of these claims, and all of them are undeniably user error.
 

Joe F

Disruption is hard.
Sep 19, 2016
1,958
8,629
Outside Philly
All that, and not a single picture of the car "penetrating the workbench"?

No dashcam video?

And "I release pressure on the accelerator which engages the brakes" isn't a thing with a Tesla, as any owner knows.

Sniff test failure?
 
  • Like
Reactions: aerodyne

MP3Mike

Well-Known Member
Feb 1, 2016
15,375
33,362
Oregon
And "I release pressure on the accelerator which engages the brakes" isn't a thing with a Tesla, as any owner knows.

If you have the car in "Hold" mode, instead of Creep or Roll mode, I think it does engage the brakes after stopping. (Only available on the Raven and newer Model Ss.)
 
Last edited:

Joe F

Disruption is hard.
Sep 19, 2016
1,958
8,629
Outside Philly
If you are in "Hold" mode, instead of Creep or Roll mode, I think it does engage the brakes after stopping. (Only available on the Raven and newer Model Ss.)
My Dec 2016 S only engages hold mode on brake (hard) depress. Newer is different? Wasn't aware of that. Thanks.

But, no pics? Very odd.
 

mtndrew1

Active Member
May 12, 2015
1,329
3,721
Gardena, CA
I’m assuming that a garage filled with such machinery has a surveillance camera of some sort? Any pics of the aftermath or video of it happening?
 

TSLA Pilot

Active Member
Mar 12, 2013
1,698
2,324
United States
While I wanted to give a fellow pilot a benefit of the doubt on this, unfortunately, I had to mostly stop reading here:

As noted numerous times in threads all over this forum in elsewhere, the motors are not powerful enough to overcome the brakes. If you've got the brake fully pressed, even if somehow the motors were trying to output full power you wouldn't go anywhere. I've tested this personally on several occasions, and it's impossible for even the most powerful version of Tesla vehicle available to overpower the friction brakes. This is by design. There's always instances where you need to stop more quickly than you could accelerate. Anyone can use the stopping distance numbers and such to calculate out the values and show that the stopping power far exceeds the vehicle's potential output power, and that's not even counting losses to the road and the like in those calculations.

I skimmed the remainder of the post...

Mainly, keep in mind that the Robert Belt paper is complete garbage and has no basis in reality. As I've noted previously, It doesn't even get close to correctly describing how Tesla's accelerator pedal sensing works at all. The setup described in that paper is something a hobbyist grade motor controller might use and has nothing to do with how Tesla's systems read this data. Without redoing my entire previous analysis (you'll have to hunt the forum, I don't have time at the moment), I'll briefly note that Tesla's system uses independent and fully isolated power supplies for the dual pedal sensors. These have a specific operating range. A value of 5V or 0V (as claimed in the paper) is just rejected outright, because this is outside the range of acceptable values for either sensor. Both sensors have to match exactly for there to be torque applied.

I know several people who have driven my Tesla vehicles and had brain farted over foot position at one point or another, thinking they were on the brake because the vehicle was slowing to a stop while they used only the accelerator pedal. There's definitely a subconscious disconnect with some folks, even in experienced drivers, over this type of control that takes some getting used to. If the car is decelerating, something in the brain sometimes tends to assume and believe your foot must be on the brake when it isn't.

In any case, I'll gladly extend my previous bet to the OP. We can schedule bringing the car and $10k in cash to my shop where we'll examine the vehicle logs together, step by step. If they show plausible user commanded acceleration, I keep your $10k and post the log analysis publicly. If they show no pedal press but power flowed and the car accelerated without user command, I'll double your $10k, post a detailed analysis that backs your position, and offer to testify to the same in whatever legal proceedings you pursue in the matter.

For the record and full disclosure, I'm undefeated on this. I'm a softy, and have thus far let everyone off the hook as far as the public shaming goes (as all have plead with me not to do this), along with not even beating them up over the cash aspect (even though I doubt I could expect the same from them had they "won"). At this point it's a waste of time for me, so I don't think I'm going to go easy on the next folks who lose this bet. I've personally examined quite a few of these claims, and all of them are undeniably user error.

Okay, we should ALL get this factoid: nearly all (>99.99%) of these events are pedal misapplication. Perhaps even this one as the brakes are at issue here. IF they were applied, in full, the car would not have been able to move. (However, I suspect that they weren't applied at full strength given the startle factor, physical limitations posed by the acceleration event, and perhaps the driver partially out of position, and a host of other potential reasons.)

But, just FYI, and especially for wk057: I have one that's not, the proverbial one-in-a-million.

I've not shared this prior as Tesla called and apologized, advising that they had, indeed, "found a problem" and addressed it via software. It's worth reading. (And wk057--feel free to contact me offline if desired.)

Driver's Personal Background: Also a professional aviator (for decades) so I'm pretty sure I can keep track of my hands and feet while operating complex machinery as that's what I'm paid to do. Also: educational background in Human Factors Engineering. Age (at event): ~50 so still mildly competent.

Vehicle: 2013 Model S 85 (RWD); early VIN (~66xx, IIRC; ~July 2013 production); this was our original Model S, of which many, many have followed.

Event: A late-2013/early-2014 INTENTIONAL and sustained full throttle ("WOT") application at highway speeds. After several seconds, our little "thrill ride" was over and I fully released the accelerator as speed was quickly approaching (or well into) 3-figures.

Was very surprised to see the power display show NO change (full power draw indicated) and the car continued to accelerate at an unchanged pace.

With that Human Factors Engineering education in my background, and having significant knowledge of these sort of events from waaay back in the Audi 5000 days (to include the Toyota/Lexus floor mat/driver panic/ineptitude fiasco*), I immediately suspected pedal entrapment by the floor mat. However, a quick glance revealed the pedal was fully up/completely away from the WOT position.

I also confirmed (by looking) that BOTH of my feet were fully off the pedals, just because I was having a hard time believing this event was occurring.

Knowing that I'd have only one chance to pull this off, I was preparing to firmly apply both feet to the brake pedal, but then realized, "Hey, why not just try 'Neutral' instead?"

So I did.

Power indication immediately dropped to flatline and the car began coasting.

Multiple attempts to repeat this event, both later that night and the next day, were unfruitful. (I suspected electromagnetic field issues as we had, for the first time ever, had both a cell phone and an iPad charging at the same time on or near the central tunnel. With low battery levels and high power draw, I suspected the ECU might be in the armrest area, thus exposing wiring between the accelerator and the ECU to EMI issues? Pure speculation on my part, but it's all I had to go with.)

Wrote an extensive and detailed report requesting Tesla do a deep dive into the car's logs, which was accomplished at some later date, on-site at the Tesla Service Center. That, in turn, lead to an eventual call from the "zone service rep" both acknowledging the problem was real, and the software correction.

All just FYI, but I do wish to point out that a little knowledge can be a dangerous thing as complex systems can fail in very complex ways.

If the event above had not occurred to me then I too would find it impossible for such an event to occur, but it did.

That said, the OP event, as described, has all the hallmarks of pedal misapplication. In the OP's defense, however, is that it has now occurred repeatedly so one would think that this would be very unlikely to be the case . . . .

Would love to have the logs reviewed.

(Is he local? If so, let's pull his logs and get to the bottom of this one ASAP.)

*Lexus Crash: An Avoidable Tragedy
 

wk057

Senior Tinkerer
Feb 23, 2014
5,659
11,430
Hickory, NC, USA
(And wk057--feel free to contact me offline if desired.)

Definitely sounds interesting and would love to chat about it. I would PM you, but I can not at the moment due to my account status here. I think others can initiate PMs with me, though, but there's some glitch with replies it seems.

Even more interesting is that software from about early '14 is where I started my reverse engineering efforts on the drive unit firmware... and I think there may have actually been some code paths that could screw up in your particular situation. This is unrelated to sudden acceleration, though, since in your situation the vehicle would have already received a true WOT event it was handling, but some miscommunication between the PM and DI modules in a particular case that I'm guessing you ran into could cause a delay in responsiveness there. (This was fully rewritten at some point in late '14, as well as additional hardware interlocks added around Q4'14.)

In your case, my guess is you abruptly released the accelerator from WOT to 0 during a particular spot in execution on the DI module, which I believe could trigger this glitch.

There's just still no code or hardware path that would cause sudden acceleration from a stop without user action, however. Maybe... MAYBE on a '12 or early '13 S someone might be able to convince me of some torque responsiveness related glitches... but even then, Tesla did the hardware and software side of this right and in a way that makes it so you're not going to get moving without proper input. Definitely some glitchiness on older motors on abrupt torque changes, but ironically it's always been more difficult in the code to gain h/w and s/w "approval" for an increase in torque. Holding and decreasing are generally less scrutinized.

If the event is repeatable, then the OP should stick a cam in the car that can see their feet and the pedals, replicate the issue, and make headlines.
 

TSLA Pilot

Active Member
Mar 12, 2013
1,698
2,324
United States
Definitely sounds interesting and would love to chat about it. I would PM you, but I can not at the moment due to my account status here. I think others can initiate PMs with me, though, but there's some glitch with replies it seems.

Even more interesting is that software from about early '14 is where I started my reverse engineering efforts on the drive unit firmware... and I think there may have actually been some code paths that could screw up in your particular situation. This is unrelated to sudden acceleration, though, since in your situation the vehicle would have already received a true WOT event it was handling, but some miscommunication between the PM and DI modules in a particular case that I'm guessing you ran into could cause a delay in responsiveness there. (This was fully rewritten at some point in late '14, as well as additional hardware interlocks added around Q4'14.)

In your case, my guess is you abruptly released the accelerator from WOT to 0 during a particular spot in execution on the DI module, which I believe could trigger this glitch.

There's just still no code or hardware path that would cause sudden acceleration from a stop without user action, however. Maybe... MAYBE on a '12 or early '13 S someone might be able to convince me of some torque responsiveness related glitches... but even then, Tesla did the hardware and software side of this right and in a way that makes it so you're not going to get moving without proper input. Definitely some glitchiness on older motors on abrupt torque changes, but ironically it's always been more difficult in the code to gain h/w and s/w "approval" for an increase in torque. Holding and decreasing are generally less scrutinized.

If the event is repeatable, then the OP should stick a cam in the car that can see their feet and the pedals, replicate the issue, and make headlines.

Very informative post and concur with your assessments. I'll check in and see if I can PM you or find you via your website, after the conf call for Q3.

However, the OP's situation is most certainly NOT the event I encountered, but has every indicator of pedal misapplication instead. Yet, the circumstances are so unique and peculiar that the OP should be running a dash cam or three, with one pointing at the pedals . . . I am open-minded enough to consider all possibilities.
 

Killerzeit

Member
Oct 4, 2016
13
4
Los Angeles
My Dec 2016 S only engages hold mode on brake (hard) depress. Newer is different? Wasn't aware of that. Thanks.

But, no pics? Very odd.

My 2015 is obviously the original way, too. I got into a few newer Teslas recently and they are true one-pedal driving. Lift up and the car decelerates enough to make it practical that it comes to a complete stop and it engages the Hold by itself, then you just accelerate to go again. It's fantastic and I wish it could be updated.
 
  • Like
Reactions: pilotSteve

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top