Unfortunately anything Musk adjacent will result in people absolutely jumping on the hate bandwagon before there are details.
Let’s wait until we know more details of what happened. This could’ve been the trucks fault, or it could’ve been the driver falling asleep and blaming the truck. If this was poor design on Tesla’s fault, then yes we criticize them, however we criticize the company as a whole (including musk), not just their CEO.
I’m not sure why everyone is so entrenched and hyper fixated one way or another on one human being.
I don't think we know the facts. Are there good forensic photos of the accident? I saw the video and it was hard to see any details other then the truck running up the driveway.
For example I'm really interested in how they determined that the "back" tires were locked.
The back axle has 2x the power of the front in the Cyberbeast. The suspicious part is the the driver said the steering was non-responsive.
Without better information, I suspect either user or system failure with full throttle and brakes causing the front wheels to lock up (because they have more brake force and half HP) causing the loss of steering and back wheels overpowering brakes.
In just about any vaguely modern car hitting the brakes will disengage the throttle. In just about any car the brakes should be able to overpower the engine. There is no situation in which a car should be moving if you hit the brakes and the gas simultaneously.
Even if this is user error, there's something clearly wrong with how the car behaved in this situation.
There seem to be barely any legal requirements for Teslas in the US. The Cybertruck as is wouldn't be legal in the EU (and hence is not sold). How the lack of indicator stalks is legal is beyond me, as is the utter lack of buttons or the lack of speed indication in a driver's line of sight.
The behaviour of using a screen with submenus to setup things while you're driving is madness if you think about it. We're not allowed to use our touch screen phones while driving but are allowed to use a touch screen bolted-on touch screen.
I’m currently renting a new Ford F-150 ICE truck and all of its UI is controlled through the enormous touch screen. There are physical buttons too, but they’re just remote inputs to the central UI. It also has a video display behind the steering wheel, where half the display is dominated by a menu, and gauges are displayed on either side of the menu. You can barely see the speedometer.
I own a Tesla Model 3, and have ranted about how awful its touchscreen only control is bad for ages, but the Ford is significantly worse. Just today, the truck’s UI bugged out in such a way that half the exterior lights were blinking, while a period bell sound would blast every few seconds. A message on the screen complained that the car “needed to be running to…” with the rest cut off by a nondescript grey square.
With regeneration you do have a desire to have power coming in from the wheels so a hard disconnect isn't desirable in all circumstances.
I know, by observation, on my Prius that "panic stop" brake pressure disconnects the hybrid system due to all the noise that the contactors make when this happens, vs. lighter braking within the regeneration limits. Braking and power never ever fight one another though. I've tried "power braking" the Prius, like you can with a normal automatic transmission car to load the engine, and it just doesn't do anything at all.
This article about the physically stuck pedals says that "Thankfully the truck cuts acceleration when you press the brake pedal..." but this seems to show otherwise.
Burns outs and drifting for the most part, assuming we're talking less then legal on public roads reasons.
On a manual ICE vehicle it makes some sense to since you need to hold some brake while raising engine RPM for until the clutch engages. Not really sure what reason there would be on an electric vehicle to command the main motors to accelerate while brakes are applied.
>We have reviewed logs and due to the terrain the accelerator may or may not disengage when the brake is depressed
Can it be clearer than that?
They literally say that the Cybertruck may continue pressing on the gas pedal while the driver is slamming the brake pedal. That's, like, the first thing they tell you not to do.
Additionally, the engine being able to overpower the brakes while stationary (or close to that) is not something that you commonly see in automobiles. As in: I'm not aware of any other vehicle with brakes designed that badly.
As a cherry on top, the steering wheel was also non-operational during the incident. Which is surely an innovation in drive-by-wire implementations.
> Which is surely an innovation in drive-by-wire implementations.
I wish it was illegal to have drive by wire in brakes or steering. These are the two most vital systems in a car that I absolutely need to retain control of in order to stay alive.
All software has some bugs, so we need to acknowledge that. Brakes and steering are too vital to let them be controlled by software alone, when we know there are always some bugs in all software.
Aside from the obvious answers (planes are regulated differently from cars, getting a pilot's license requires more training, etc), let me point out that autopilots for planes existed for over a century, but we are only getting some semblance of that in cars now because flying is a much easier control problem to solve.
Unsure why mention autopilot when talking fly-by-wire. Yes it won’t be as redundant as airplane fly-by-wire, but merely by putting miles and manufacturing hundreds of thousands copies one improves quality.
Yeah I mean I'm all for banning people from driving an Airbus on public roads. And Boeings, Bombardiers, Leers, and every other FBW airplane. We should probably extend this to all planes though, just to be safe.
Am I the only one that thinks it's weird that Tesla (and most modern cars, I guess) is the only one with access to these log files? Is there any independent verification of authenticity of these files?
>Am I the only one that thinks it's weird that Tesla (and most modern cars, I guess) is the only one with access to these log files? Is there any independent verification of authenticity of these files?
No, you're not the only one.
These should be regulated like black boxes on aircraft, and mandated to be available to the NHTSA.
If you've ever driven a vehicle with air brakes, they are even more "grabby"; to the extent that you need to be careful not to cause the tires to lock up.
There are lots of ICE vehicles that have enough power to overcome their brakes
My diesel pickup could easily rip the brakes to shreds under acceleration. Electric vehicles are known for crazy amounts of torque as well… sounds like somehow this guy had both the accelerator and the brake pedals depressed at once. It’s weird
> There are lots of ICE vehicles that have enough power to overcome their brakes My diesel pickup could easily rip the brakes to shreds under acceleration.
I'll dispute that. Have you actually tried that? If so, your brakes are defective.
There's an easy intuitive way to visualize this. Remember that any change in speed or direction is acceleration. Braking is just acceleration with a negative sign.
The engine generates horsepower to accelerate the car from zero to some speed. While we don't usually speak of HP when talking about brakes, it is equivalent. We can also compute how many HP the brakes are producing when accelerating (with a negative sign) the car from that speed down to zero.
So: a car that has a more powerful engine than brakes would be able to accelerate from 0 to 60 (or pick some speed) faster than the time it takes to brake from 60 to zero.
Can you think of any car or truck that can accelerate faster than it can brake? Doubt it (unless the brakes are broken).
To your truck example, I have a one-ton turbo diesel truck. I hadn't tried this experiment so was curious after your comment and went for a drive. As I expected, if I hold down the brakes the truck won't budge even at full throttle. The brakes are massively more powerful than the engine.
They have to be, after all those brakes are designed to bring the full rated capacity of the truck and trailer (28,000lb) to a stop from highway speeds. The engine, while very powerful, has way less power (acceleration potential) than the brakes.
"2024 Tesla Cybertruck Dual Motor Foundation Series Specifications
0-60 MPH 3.8 sec
BRAKING, 60-0 MPH 126 ft"
Car accelerated ~~2/3 of the way on the video~~
Edit after re-watching: car was floored downhill with slight wheelspin for 8 seconds, it was slowing down when you can hear screaming tires for less than 2 seconds. Looks to me like it was fully braking after all, after someone decided to floor brand new 4 second to 60 EV vehicle into a building and shat his pants.
It just occurred to me that this should be possible in a front wheel drive vehicle with the parking brake applied but for some reason the rear wheels still roll if you forget to release it. Not something I’ve ever tested on purpose though.
But video proves no claims. No skid marks, no locked steering. Looks like he just ripped too fast and didn't have enough distance to brake.
> mods nerfed this submission into the ground
Because it's Jalopnik, known anti-EV and especially anti-Tesla page ran by Gawker. Look at all the submissions from the domain and tell me they aren't biased.
I think, if someone is holding down both the accellerator and brakes, there is only one safe resolution of that dillemma. And it’s the one that doesn’t involve the car turning into a ballistic projectile.
People will be people, so any other option seems like the height of arrogance.
I could be misremembering, but I think I remember Teslas having a feature that would apply slight torque to the motors when braking on hills to prevent the car from rolling backwards when switching from brakes to the accelerator.
I wonder if this was a malfunction of the system that detects the car being on a slope which caused it to accelerate forward despite the brake being pressed?
There might be some conditions where it didn't work exactly like it should, i.e. when I enter mine and press brake before car booted up and try shift into drive it says press brake first, even when it's pressed already.
That's honestly probably intentional. The car wants to see the brake in both states before it'll boot. If it wasn't quite paying attention (because e.g. it takes a second to start the controller in charge of this stuff) it might've only seen it in the pressed state.
It's unclear to me from the tweets if anyone confirmed the driver removed his foot from the accelerator. I'd absolutely love to dump on Tesla, but people often do stupid things in panic situations.
"due to the terrain the accelerator may or may not disengage when the brake is depressed"
There might've been some amount of driver error involved, sure, but that statement sends shivers down my spine. Either of "the accelerator always/never disengages when the brake is depressed" is sane. The idea that it may or may not, depending on external factors outside the driver's control, is just plain terrifying.
My car's accelerator doesn't disengage when the breaks are pressed. I've never owned a vehicle that behaves like that. If my car was smart enough to sometimes disengage the throttle when I have then break also engaged I can't see a case where this would be less safe than never doing it. If I was intentionally trying to wreak my car or do some weird stunt by simultaneously engaging the break and throttle and the car sometimes tried to stop me I guess it would be annoying.
ICE cars are designed such that the brakes can always overpower the engine. If you put both pedals on the wall the car will not move.
Because electric cars have phenomenal torque, it's not possible to make brakes big enough to stop the car against full application of the accelerator. So they have to have disconnects.
I had to fix the emergency brake on my car once because it wouldn't pass the annual safety inspection in my state. IIRC the way it's tested is they engage the brake and then put the car in drive. If it moves under . . . I think it was something like 3k rpm, the car fails the safety check.
> IIRC the way it's tested is they engage the brake and then put the car in drive. If it moves under . . . I think it was something like 3k rpm, the car fails the safety check.
This test won't work at all for a FWD car. The parking brake locks the rear wheels but the front tires are free to rotate under engine power and since on a FWD car most of the weight is on the front axle, the car will happily slide the ligthly loaded locked rear wheels along for the ride.
Traditional ICE cars with auto transmissions allow you to do this, by design.
But you’d have to have a front wheel drive car and engage the parking brake / handbrake to lock the rear wheels like that.
Which is not possible in a Tesla.
For a Tesla to have locked the rear wheels whilst the front wheels continue to drag the car forward, does, on the surface seem like a failure in software or electronics somewhere.
I should test what happens in my Tesla :-)
(Side note: I wish Musk would just leave Tesla so we can all feel less shitty about owning them :-/ )
It's unclear to me which axle even locked up. How would the driver even know? The witness marks would be flat spots, but no photo evidence is given. For all I know, the skid marks are from rear tires burning out.
The article claims the driver mentioned "unresponsive steering", but that doesn't make much sense as the vehicle pretty clearly follows the road (to the right). It's not like one would turn at the bottom of the dip (into the ditch).
I can't blame anyone with such little information.
The other thing I also think about in these cases is that unlike an automatic transmission car, the regenerative breaking slows the car the same as light break application, so in regular driving you don't use it as much.
Having driven a many different cars, I suspect without as much experience with manual transmissions or regenerative brakes, it is much more likely for a driver to panic stomp the gas.
Indeed "Pedal Misapplication" is in fact a routine term of art[1]. Drivers hit the wrong pedal all the time when they get surprised. But of course this accident must be because it's a Tesla.
> Drivers hit the wrong pedal all the time when they get surprised.
Article says rear wheels were locked up, which implies heavy brake pressure.
Why the fronts weren't also locked up is an open question, but in any properly engineered car that should be impossible. The front brakes are always larger and more powerful, so if enough braking force is being applied to lock the rears, the front wheels should also.
Article's source is two tweets. Let's maybe consult Occam before diving too far down the conspiracy hole. Confirmed Priors are a hell of a drug, as it were.
The point being that there is a extremely simple explanation for this accident that people are willfully ignoring because of the manufacturer of the vehicle.
I just did. The tweeter is a two-pedal driver who leaned on the accelerator by accident, didn't realize it, and stomped on the brakes to stop the car. Then fibbed about it a little on Twitter for clicks. Why do you believe this one twitter user over me? Because one confirms your priors and the other doesn't.
> Why did it keep driving forward with only one axle locked up by the brakes?
Again, you're citing facts not really in evidence. If you want to believe a two-tweet conspiracy without confirmation, I can't stop you. I'm saying that it's probably not the truth and that a routine loss of control is a better explanation.
Stomp on both, car breaks hard by locking all tires. Blame "Tesla bug" because it didnt stop in 50 feet ... after you sped up by max accelerating for 8 seconds flat to ~80mph.
Because you then deliberately stomp the brakes to stop the car you forgot you commanded to accelerate. Again, this happens all the time, and no one cares when it's a grandpa in a F-150.
Sure, but the driver doesn't have separate pedals for front and rear brakes. GP was claiming it might be a driver error, but in the absence of separate brake pedals for front and rear, it is not an error the driver can make.
The more electronics and software involved, the more errors like that will occur. I expect it to worsen as the car gets older, with fewer updates and available parts. The only exception is strict testing, similar to what is done with commercial airplanes.
> Drive by wire throttle is in the vast majority of cars made in the last twenty years because emissions and fuel efficiency require it.
Electronic throttle controls is still safe because if the code hits a bug and the throttle becomes uncontrollable you can still use the brakes to stop the car.
But if you also have drive by wire braking system, you can have a big problem.
I agree, but none of them has autopilot, and the article lacks a technical assessment of why it happened, but I reckon it was due to that. ECUs, on the other hand, don't decide their own actions; they are essentially microcontrollers. All modern cars have plenty of ECUs, each for specific functions, like terrain ECU, brakes ECU, etc. They are extensively tested for FMEA (Failure Mode and Effects Analysis), so the margin of error is slim compared to modern electric cars with all these autopilot technologies.
>Likewise, applying the brake pedal simultaneously with the accelerator pedal will override the accelerator pedal input and cut off motor torque, and regardless of the torque, sustained braking will stop the car.
This is quite a stronger statement than: "the accelerator may or may not disengage when the brake is depressed". The former statement is true in my experience with my Model 3: if you press the brake and the accelerator at once, you get a notice on the screen saying that it's not going to let you be an idiot. I wonder why the Cybertruck (seemingly) doesn't behave the same way.
More evidence about the failure of engineering standards at Tesla. It's actually very sad - they genuinely seemed to be doing great stuff before (regardless of manufacturing issues).
As an aside - it's hard enough being a bicyclist in the US with all these absurd-ly large trucks prowling the streets whose drivers can't see bicyclists either in front of their hoods or in their side mirrors. Now we have this moronic vehicle which adds in faulty controls and sharp pointy body panels that'll slice through you when you hit them.
On a psychological level, I fear there's something deeply wrong in the US as they lose their global hegemony. I respect the second-amendment and all that, but what civilized country markets a vehicle as being "civil war" ready and "bullet proof" ? Sheer insanity.
AFAIK the brakes on most if not all vehicles are more powerful than the engine, and the front ones more so than the rear. To have the rear brakes locked up while the front gets presumably overpowered by the motor is a very odd situation, perhaps best described as a "reverse burnout".
Not sure how regenerative braking works if it doesn't disengage the accelerator when the brakes are pressed. Yes in this case the brakepads would be used, but I'd also expect the motor to be off.
Are you saying the car should go if you press the accelerator while you're holding the brake? No car I've ever driven worked that way, if I'm holding on the brake and I apply pressure to the accelerator the car stays still.
Right - it's how Audi's "pseudo" launch control works. Push the brake pedal to the floor. Push the accelerator to the floor, and then push it past a last detente while the ECU holds it at 3500rpm and builds up the turbo boost, and then you can release the brake.
You can feel that the brakes are what's holding you, but never do you feel that they're about to fail and you're about to floor ~520hp unexpectedly.
They're accomplished by having the more powerful front brakes hold the car while the less powerful rear brakes allow the rear wheels to spin. Sometimes water is used to reduce traction. Sometimes the brakes are modified so that the front axle can be controlled by a hand lever without applying the rear brakes.
In this case the Tesla was able to overpower its more powerful brakes and drag the car forward. That should never happen.
Update: hijacking my own comment to say that the HN mods have nerfed this story submission into the ground: https://hnrankings.info/40782552/
I had a 1980's car in which the ECU had a sense line to the brake light circuit, and if the brakes were applied, engine power was instantly, and dramatically, cut.
Every drive-by-wire-throttle car I've ever driven - and DBWT cars have been around since the late nineties, do the same thing - press the brake, the ECU goes to idle immediately - sometimes in sports cars it's after a few seconds. There is zero reason for this pavement queen to be set up to permit left-foot-braking under throttle, especially when it is so heavy and powerful.
The federal government and EU regulators need to find some balls, fast, and reign in Musk. Every single idea he's had to "revolutionize" cars - beyond going electric, not really his 'idea' - has been asinine. In some cases he's pioneered Stupid and the auto industry shrugged and followed right along because regulators let him:
* going backwards at least twenty years in pedestrian crash safety by making a flat-nosed vehicle whose body panels can deflect a sledgehammer blow, instead of crumpling when hit by a person's body
* self-driving and lane-keeping cruise control with no radar has caused numerous crashes (thanks to "phantom braking" when the car freaks out and thinks something is in the road) and when the car doesn't see something because of dirt, sunlight, or water on the cameras - has killed dozens (by low estimate) of people...both occupants, law enforcement, and the public and that's just the well-publicized cases. I'm betting the actual numbers are more like thousands. Dual camera systems simply are not sufficient because even the very best pro photo digital cameras are still worse than the human eye in resolution and dynamic range.
* yolk steering wheels, though at least they kinda sorta addressed that with the variable ratio steering in the CT. There is an astounding lack of hubris to look at a fucking steering wheel and say "I can do better!" - especially when "better" has been tried before at various times and was a miserable failures. Musk looked at F1/Indy and said to himself "SEE? BETTER!" and forgot that those are being used by people who are at the absolute pinnacle of driving skill and rarely have to do anything approaching a sharp turn.
* no indicator stalks...putting the most commonly used control on the car is now a moving target and can have any rotation
* vehicle controls on a screen that depending on the Tesla model, anything on the road is impossible to see while you're looking at the control, which you have to do because there are no physical controls or even haptics.
Did I leave anything out?
In short:
- No self/autonomous vehicle tech without radar and some sort of LIDAR to supplement cameras
- No steering yolks
- Sheet metal must crumple appropriately in a collision with another car or a pedestrian
- Indicator stalks required
- Speed and basic indicators must be in front of the driver. Want to do it via a HUD? Fine.
TLDR after watching Video clip: "I accelerated my 3 ton 4 seconds to 60 EV tank downhill for 8 seconds and was shocked it didnt stop in the 2 seconds I was actually applying brakes"
It was easily doing 60 at the bottom, but kept accelerating uphill for couple more seconds, probably to 80 before he started braking. No car would stop in those 50 feet. It wouldnt make it even if he started braking just after crossing the bottom.
Jalopnik is a pretty well known Tesla FUD website. Not sure how I'm the first one to point this out. Take everything on that site with a grain of salt.
The video is literally in the article. Seems like your the one with the bias by being unwilling to read the entire Jalopnik article.
You won't see me putting my family in a Tesla and this article certainly isn't the only reason why. Plenty of similar incidents to choose from. I prefer not to get beheaded by my car because it decides to drive through a semi at 70mph.
The majority of times I see critical articles and news being labeled as "FUD" is in crypto scams and memestock communities, which is very telling about the state of Elon's stock. Though I must say, it's fitting.
The rear brakes locked and were dragged ~50 feet. The front axle CONTINUED pulling the Cybertruck for those 50 feet, with the brakes applied.
There is absolutely no engineering case in which this should be expected behavior.