- cross-posted to:
- [email protected]
- cross-posted to:
- [email protected]
The electric car manufacturer Tesla had to issue a massive recall this month to fix faulty hood latches that can open while its cars are driving. The problem affects more than 1.8 million cars, which means it’s slightly smaller than the recall in December that applied to more than 2 million Teslas.
The problem, according to the official National Highway Traffic Safety Administration’s Part 573 safety recall report, affects model year 2021–2024 Model 3s (built between September 21, 2020, and June 2, 2024), model year 2021–2024 Model Ss (built between January 26, 2021, and July 15, 2024), model year 2021–2024 Model Xs (built between August 18, 2021, and July 15, 2024), and model year 2020–2024 Model Ys (built between January 9, 2020, and July 15, 2024).
The problem first became apparent to Tesla in March of this year after complaints about unintended hood opening from Chinese customers. By April, it had identified the problem as deformation of the hood latch switch, “which could prevent the customer from being notified about an open hood state.”
Although the problem is with the hood latch, as with many Tesla safety recalls, the problem can be fixed with an over-the-air software patch. The new software is able to detect if the hood is open and, if so, will display a warning to the driver to alert them to stop their vehicle and secure the hood.
You must log in or register to comment.
The new software is able to detect if the hood is open and, if so, will display a warning to the driver to alert them to stop their vehicle and secure the hood
This should not be legal. They should be forced to recall vehicles and replace the faulty part instead of kindly asking drivers to pull over when the part fails.
The shit this company gets away with is astounding.
Driving on the highway
Hood pops open. Can’t see anything. Try to brake but crash.
⚠️ Warning! Your hood is open. Please pull over in a safe location and secure your hood.
Tesla: OTA update successful
Yup. No reason to rewrite the playbook.
“Full self driving” detects an imminent collision of it’s own doing. Car beeps and shuts off “full self driving”
Human was “in control” at the time of the crash, not our fault.
Human was supposed to be in control the whole time. In a Tesla and in basically any car, very limited exceptions apply. If some asshole goes to sleep at the wheel that’s the asshole’s fault.
Yeah but turning off assists the split second before a crash is a bullshit way to try and claim it wasn’t involved in the crash. Tesla was caught doing that a couple years back.
Yeah, that’s just fucking lying to the public.
This is an unrealistic standard. We are not robots. We didn’t come with a standby mode unless we’ve actually worked on silently watching a vista while being ready to act at a moment’s notice. And honestly, even the people we pay to do that have problems with inattentiveness and falling asleep.
We need to jump from actually controlling the vehicle to a self driving system good enough to take a nap in. The middle part is extremely dangerous to humans.
I disagree, even the middle part is safer than before. It’s a net positive and will continue to improve.
It’s safer as a backup system yes. As a nudge in the wheel near the lines, as a light brake before the AEB has to kick in. Tesla’s is not safe. It makes people look elsewhere for engagement when the car is not ready to take full control.
This is an unrealistic standard. We are not robots. We didn’t come with a standby mode
True. That’s why driving with assist sytems does not make driving more safe.
We need to jump from actually controlling the vehicle to a self driving system
Get a Mercedes. They guarantee you some (limited, but still) hands-free time in the driver seat.
Just like the rumored cut all driver aids instant before crash to say it wasn’t due to any of the auto pilot features
The only way the hood can pop open on the highway is if it was open before you departed, so the warning would alert the user just like the switch did before they can drive to a dangerous speed.
It’s not the actual latch that’s faulty, but the warning the driver should get, if they haven’t closed the frunk properly.
does tesla just not bother to hire a qa team or something?
It’s pretty tough to say without looking at comparable recall rates for other companies. My Toyota Corolla had two recalls that I know about while I owned it, and Toyota is known for their reliability. Tesla is just always in the news because they’re always in the news.
Toyota is also known for their reluctance to issue recalls even though they REALLY need to do a recall because they don’t want to tarnish their reputation for reliability. Often waiting until the government forces them to issue the recall. I’d much rather the Ford approach and just issue recalls like candy on Halloween. Sure having 700 recalls sucks, but driving a defective car sucks even more.
Cheaper to just let a bunch of people get injured/killed
Not always. But yes, there is a price.
Microswitch lever fatigue is what this sounds like and it’s really not the kind of thing that a QA team could ever detect without years of testing. This is just how it’ll go as we add more bells and whistles to all our cars. More obscure new issues will be identified years down the line and added to institutional knowledge for future use.
To be fair, it’s not an issue with just Tesla, but basically all modern software. The end user has become the beta tester.
This isn’t software, it’s a car. It’s highly regulated. NHTSA doesn’t care if it’s a software issue.
deleted by creator
Recalls aren’t uncommon. You just don’t hear about most because it’s not trendy.
One of my vehicles is at risk of catching fire. The other is at risk of its axle falling off.
These are major brands, within the past 5 years.
The latch is fault, and so is the sensor. Sensor doesn’t go off when the latch starts to fail from deformation.
According to the description, it’s just the sensor, not the latch. The microswitch has a lever like many do and that lever can become bent if damaged which would prevent it from warning the user if they failed to latch the hood. Most older cars just had a secondary latch so if you failed to latch it completely, at least the secondary one would catch it…
Although the problem is with the hood latch, as with many Tesla safety recalls, the problem can be fixed with an over-the-air software patch.
Not sure what description you’re talking about, but I’m basing this on the article itself.
Even in your reply…you say the problem is the actual latch being physically damaged.
It’s the owner that’s faulty for buying one in the first place.
I mean, this can happen with any car that has a hinged hood (so nearly all cars)
Is a frunk a front trunk, or is that a typo?
Front trunk, yeah
This right here is why there should be a distinction between software updates and physical recalls. Calling this a recall without actually taking the product in and fixing the product is really deceptive.
It should be a traditional recall though.
Yeah, but that’s how it works according to the NHTSA.
That’s not what they’re saying. It’s essentially a “door ajar” warning. The sensor is what’s failing, rather than the physical part.
Isn’t a sensor… A physical part???
Not if it can be fixed with an update, that’s a software issue.
But it’s not. If you read the article it literally states it’s due to not closing because of deformation of the hood latch switch. Which is a physical object. Sure, you can get a notification now that the thing is open, but guess what, if it flops open while driving, I’ll probably already know that.
And if you’d read it, you’d notice they specifically say it’s the hood latch switch, and not the latch itself, that is deformed. It doesn’t pop open on its own, it falsely detects that the hood is properly closed when it actually isn’t, so it doesn’t warn the driver to go close it before driving off.
There’s no need to do a physical recall if Tesla can bodge some other way to detect the hood is still open and tell the driver to close it.
Each country has it’s own authorities. Hopefully they don’t get away everywhere…
Oh so here’s a fun thing. All American corporations have this level of freedom. We’re just paying extra attention to Tesla because their CEO can’t keep himself out of politics and the news. Ever notice you only see the CEO of GM/Stellantis/Ford when it’s a crisis or a new CEO? That’s how it works in a functional business. They aren’t any less shady, they’re just better at brand and scandal management.
Ford did this. On multiple years of some vehicles the fuel injector can crack and leak fuel onto the engine and lead to a fire. Their fix is to put in a fucking drain tube to drain the fuel away if that happens rather than replace the faulty part. I’m wondering if there are any legal options to make them just replace the part rather than their half-ass non-fix.
Not even.
Ford’s fix required them to physically add parts.
This is more like if Ford just wrote a software update to detect the crack and leak, then pop up a warning that you need to pull over and “secure” the fuel.
Using software to patch design flaws seems to be a recurring cost-cutting pattern these days. Look at the MCAS of the Boeing 737 Max. This is how civilizations go to shit.
deleted by creator
I would say more annoying is to getting those alerts more and more frequently. According to the article this is a software “fix” for the lock banding and stopping functioning.
deleted by creator
It literally states the program is due to a deformation of the switch, a physical issue.
I’m sure there are ways to safely engineer around it. I don’t trust any American manufacturer to do so. What guarantee do we have that it won’t continue deforming further after the fix? What happens to the patched sensor software if it does occur?
Do you unironically trust Musk enough to not open your vehicles hood at 80mph?
Really not sure why so many people on this post are having trouble understanding that concept. Maybe they want to justify their $80k paperweight lmao.
deleted by creator
Sometimes you can.
It’s common to use a Hall effect sensor for positioning. It gives off an analog value. You might be able adjust the signal threshold that you consider to be “open” or “closed” in software.
Further, this is probably something that you just don’t spend a bunch of time engineering. Pick a value that’s well with your tolerance range and move onto harder problems. When a problem comes up, you can fine turn the range.
deleted by creator
giving elon another $55b bonus should fix it
We need more children from him, because apparently we don’t have enough!
Considering it seems all his children hate him, more can’t hurt.
Yeah I was about to write that some of the children seemed to turn out sane.
Thankfully you don’t need to be one of his kids to hate him.
Do your thing, ma-gic man! Chanted the crowd as Elon in robes entered the golden restroom with his xPhone in the right and a ceremonial vial of coke in the left hand. It was followed by a series of thunder-like farts and sniffs.
Although the problem is with the hood latch, as with many Tesla safety recalls, the problem can be fixed with an over-the-air software patch. The new software is able to detect if the hood is open and, if so, will display a warning to the driver to alert them to stop their vehicle and secure the hood.
Patching the software isn’t a “fix”. Changing out the hood latch so it doesn’t come open while driving is a fix.
"Warning: wheels falling off imminent! Please pull over!’
You have 3 seconds to comply. 3. 2. 1. Goodbye!
I’m guessing this is an issue because of the increased usage of the “frunk”.
Nearly every ICE can suffer the exact same issue.
Can? Maybe.
But it’s far less likely to because you’re not using it anywhere near as much. It’s not reasonably expected use. Opening a storage compartment/froot/frunk is expected.
A 24 pin port on your PC motherboard that’s rated for 100 insertions would be completely fine.
A USB-C cable wouldn’t be.
The frunk issue is a good point, and that I hadn’t considered - thank you!
Trying to solve a hardware problem with a software solution. Where have I seen that before?
Right? It’s deformation of the hood latch. So a physical change. How long until the warning is permanent?
Thought you had made an Intel reference.
Yes, because Boeing is the only time software has attempted to solve a hardware issue…
I too enjoy taking comments out of context so I can criticize them.
How is that taking it out of context? Software is used to fix hardware defects all the time, everywhere. Just because Boeing did it poorly, doesn’t mean it’s inherently dangerous or a bad idea.
They did not state or imply that Boeing was the only company to fix a hardware issue via a software update, nor did they state or imply that doing so is an inherently dangerous or bad idea.
The verifiable context was simply that Boeing, a disreputable company, also attempted to correct a hardware issue with a software update. One might infer that the OP could be suggesting this instance may be bad given Boeing’s failure at doing so combined with Tesla’s dubious reputation due to its association Elon Musk, a demonstrably unethical person, and its record of vehicular build quality issues. Claiming the comment suggests ALL efforts to do so are INHERENTLY bad or dangerous cannot be supported without additional information from the OP, though, so criticisms to that effect rely on pure speculation.
How is Boeings failure related to Tesla, a company known for terrible hardware quality control and amazing software?
You’d have to ask op, we can only guess at anything beyond what they wrote. That’s my entire point.
And my point is that OP did try to imply that using software for that is a bad idea, I can see no other possible explanation.
Is this a real recall or an OTA update?
I hate Tesla, but a lot of news outlets are like
TESLA RECALLS BAJILLION CARS And what they really mean is they released an OTA update to fix some extremely rare race condition.
The issue is still bad, but I feel like the news outlets are sensationalizing to the point of dishonesty sometimes.
To be clear I’m not sure I understand the actual underlying issue here, so idk how deserved the headline is, but whenever I see them, I’m immediately skeptical
The problem is with a shitty latch: the hood appears closed, but it’s not.
The OTA Update doesn’t fix the shitty mechanical latch - it still doesn’t latch consistently. What it fixes is another poor design choice: evidently, the car has sensors that can tell if the hood was closed correctly or not, but this was never turned on/programmed? The OTA Updates this so now the car can warn you when the shitty latch fails.
Or who knows, maybe they initially turned off that sensor because it was going off all the time because of the latch…
Although the problem is with the hood latch, as with many Tesla safety recalls, the problem can be fixed with an over-the-air software patch.
The problem is a sensor that detects if the latch is closed. How did an OTA fix this?
It hard to tell how big of a concern this wasThey probably only used the sensor while unlatched, now they use it while supposedly latched too.
Did you not even read the summary?
“all those words I have to scroll past before commenting? No, why would I read any of that?”.
-opI did and it doesn’t really give complete answers.
The latch is fine, but the sensor to detect when the driver didn’t close the hood can break. I dont understand how an OTA can fix that, and idk how common the problem was, so I can’t tell how sensational the headline is
The issue is a faulty latch that deforms and allows the hood to open, and their fix is to have a warning light when the hood opens.
They’re not fixing the problem.
Reading it, the latch doesn’t deform, the latch sensor deforms. It doesn’t allow the closed hood to open, it fails to alert the driver that the hood wasn’t closed.
But if the sensor is broken, idk how an OTA can detect that the hood is open.
Beyond any issues with the owner of the company, these cars have multiple dangerous issues.
You cannot treat a company that makes physical stuff that can endanger lives the same way you treat a software company that makes a leisure activity platform.
Iterative design for a purely software environment is way more forgiving than iterative design for physical hardware or even software that interacts with physical hardware. You can profoundly fuck up the backend for a website and take the whole thing down until you could roll back to last known good production, you won’t kill anyone, but you’ll make the line go down temporarily.
If you profoundly fuck up an iteration on an embedded vehicle system and don’t catch it because you don’t respect safety regulation or existing engineering norms you can and will kill people.
I wanted a Tesla as a teenager, I recently realized that they are getting so low I could in theory finacance a used one, but they are so much shit lmao, worse than a Kia
I’m a school bus driver and middle school boys go apeshit when they see a Tesla - which is unfortunately often because they’re pretty common these days. They explode when they see a fucking cybertruck, which is fortunately still not often.
The good part about Teslas is how advanced their FSD is, plus their superchargers. The bad part is that they have terrible quality control, you’re driving a touchscreen on wheels, and you’re finding Elon
https://www.kbb.com/kia/recall/
Every manufacturer has recalls, like that fun Telluride one that lights the car on fire if you adjust the driver seat too much!
Kia’s are not great, but Tesla seems worse
I didn’t like it when the first one came out and I was a teenager. But, to be honest, I was very pessimistic about the transition from late 00s to early 10s in everything around me. EDIT: Switched to Linux then too.
I feel like I would get a used tesla entirely to strip it down into a Mad Max mobile.
That would be neat, they can go pretty fast too like that: you could likely find a wrecked salvage one for the project at a junk yard. As long as the battery, motor, and it’s drivable.
Id also have to learn how to circumnavigate the fucking tablet it uses but worst case scenario Id hardwire switches.
The problem first became apparent to Tesla in March of this year after complaints about unintended hood opening from Chinese customers. By April, it had identified the problem as deformation of the hood latch switch, “which could prevent the customer from being notified about an open hood state.”
Given that China is now an electric car superpower, this situation will not bode well for Tesla in that country.
I haven’t met a techie that likes the idea of Chinese made BMS either. I suspect that when Chinese made EVs come over, they’ll be a honeymoon period like what happened with Tesla before we start hearing about battery fires in people’s garages. If you or anyone you know gets a Chinese(or any really) EV, store it outside or in a detached garage/carport.
By April, it had identified the problem as deformation of the hood latch switch, “which could prevent the customer from being notified about an open hood state.”
I think there was a scandal before with the logic of that thing being not good at all and it becoming dangerous if you’ve put sufficiently heavy pressure to prevent it from closing a few times.
the problem can be fixed with an over-the-air software patch
I was about to make a bad joke about Tesla trying to fix a mechanical issue with a software patch but apparently Elon beat me to the punchline.
The comments read like a lot of people don’t quite understand the issue.
The bonnet (hood if you insist) latch may not warn a driver if it isn’t secured correctly. If it is secured correctly then it is fine. So it isn’t going to suddenly open.
If the latch isn’t shut correctly and then the sensor doesn’t report this then the bonnet may open unexpectedly.
If they can use a software update to correct the reporting then that’s it fixed.
There’s no issue with the actual latching mechanism. It’s just the sensor for reporting the latching state.
It may be that it currently works on a two value system. i.e a value for correctly latched and a value for not latched. If that’s the case and isn’t just not providing the second valve correctly then a simple software change to only use the latched value would fix this. As any other value or the absence of a value will report it at unlatched.
It’s a soft close latch, the frunk pulls the lid into the latching mechanism. The mechanism isn’t doing it’s job and needs to be replaced either with a properly adjusted soft close mechanism that grabs the lid, or with a non soft close standard latch that is very obvious to the user when it has not been properly closed.
Nope it’s the latch switch. So something that is switched when the latch is closed. Not the latch itself.
It’s a bit ambiguous, but I read it as saying it isn’t the mechanism that’s at fault, it’s user error that’s not currently being properly detected.
I don’t think it’s ‘user error’ exactly. Maybe when this has occurred, something in the frunk has obstructed the closing of the hood so it almost latched, but the deformed switch is detecting it as closed. I think they might be adjusting the switch sensitivity in software (maybe it uses a Hall effect sensor and a magnet?) so that this almost-closed condition will be reported as just being open.
The comments read like a lot of people don’t quite understand the issue…There’s no issue with the actual latching mechanism.
…“Although the problem is with the hood latch” <— literally from the article. Care to re-read?
It’s just the sensor for reporting the latching state.
You skipped over the part where a) the latch is deforming, and as a result of that deformation b) the sensor can’t detect that it’s not closed, and so c) Tesla is pushing an update that lets people know their deformed latch isn’t closed properly.
But yes, we all misread the article. Not you. Definitely not you.
And what is the next word after the bit you have quoted?
Is it by any chance switch.
The full quote is deformation of the hood latch switch. Not the hood latch.
Thanks for further confirming my point that you’re not reading it correctly.
No, it’s quite literally not. Click the article, read the article all the way. Including the last paragraph. Where my quote is from.
Then read the recall. Then lookup the part. See what it is? Oh, it’s the entire latch assembly. Good job! Proud of you sweetie. 😘 Keep licking those musky boots!
Did you read the recall? Again it says hood latch switch deformation.
That may be part of the hood latch assembly but again at no point does it say that the latch not latching is the issue. Only the reporting of the latching state.
You’re really rather pathetic and I’m certainly no fan of Tesla or Musk. A brief check of my previous posts would confirm this.
As you’re obviously not very good at reading or understanding things then that fact probably did slip by you. You seem to be only capable of latching (you might not see what I did there being a bit dense) onto certain words without understanding the full issue.
You missed the part where the latch is deforming, causing it to not close or alert the driver. The software fix is yet another attempt to dodge the fact that they do not have enough repair capacity or financial reserves for a major fleet recall.
Read it again. It’s deformation of the hood latch switch. Not the hood latch.
Thanks for further confirming my point that you’re not reading it correctly
Too bad the software isn’t open source.
diff --git a/hood.js b/hood.js - if(false) { + if(true || false) { + alert("Check your hood")+ if(crashed) { + alert(e) + } else { + load_ad("vote_for_trump") + }
While that may “solve” the issue, it’s still due to a faulty physical component. A software update is just a bandaid over the real problem.
Every day I’m reminded how glad I am that I changed my mind about buying into the early run of the 3 series. Shit quality and even a bigger POS in the head office.
deleted by creator
Who knew Tommy Boy drove a Tesla?
Surely this was all worth the 47 billion
