US investigation of Tesla steering problems is now one step closer to a recall | Hacker News Reader

US investigation of Tesla steering problems is now one step closer to a recall

apnews.com

96 pointspg_123472 comments

One thing about self driving cars is they need the ability to command the electric power steering system. I worked in EPAS for 6 years and it was a big deal when customers started wanting to do that due to liability. My guess - and it's only a guess - would be that some situation is leading to normal epas operation getting disabled but self driving not enabled. Not sure whose steering gear they use.

My understanding of the Tesla EPS system (this is also a guess, based on discussion observed between Comma 3X devs working on Tesla integration) is that there's not a separate mode it operates in while under FSD is enabled. That's in part how the safety of the EPS is validated independent of the LKAS - driver should always be able to easily overpower it. Contrast this with other car models with a "parking mode" for their EPS, which operate under greatly increased torque limits.

Clearly something is going wrong if drivers have to fight hard against the EPS, but FSD shouldn't be able to do anything that a typical LKAS can't either, safety guardrails included.

(also, thanks for your work on solvespace, great piece of software)

The wheel in FSD is hard to take control over and when you do it goes absolutely slack leading to an overcorrection. Even after having my car for years it still does a wiggle when this happens. Not sure if that’s what you’re saying shouldn’t happen. Doesn’t happen with my CRVs LKAS

OK, so this is definitely something to follow up on. But also, that other recall for the warning lights being too small that the article mentions really caught my eye. Seems like Tesla failed the brown M&M test[1] on that one.

1: https://www.snopes.com/fact-check/brown-out/

A few times I've seen a message in my car (I have an "S" and a "Plaid", neither of which are in this recall) that the power steering may not be operating properly. But despite the message, it never felt any different. The message just disappeared and I never worried about it after that...

> Also emerging on Friday, Tesla is recalling nearly all of the vehicles it has sold in the U.S. because some warning lights on the instrument panel are too small.

What warning lights on the instrument panel? Is this just a software update to the LCD display?

Warning lights/icons on the display being too small/subtle seems very on-brand for Tesla. Musk was willing to just eat fines over the aesthetics of the airbag warning on sun visors (which I admit I personally despise).

One thing that bugs me is "red" appearing on the dashboard for things that aren't warnings -- like album or podcast cover art. I still get distracted when I see something red show up, and then look to realize it's the "Commentary" podcast logo, or something like that. I think that red should be reserved for something that needs immediate attention, and that album cover art shouldn't be on the dash.

> which I admit I personally despise

That's extremely funny. I don't even notice them, but my dad ruined his sun visor once trying to get those damned stickers off, hahaha.

> Is this just a software update

Yes, most likely - but "recall for every vehicle sold in the U.S" gets you more clicks.

Recall in this context means the government (NHSTA usually) has decided a fix is required. OTA or in-person doesn’t matter - what matters is the legal obligation of the manufacturer to provide a timely safety update.

Recall is an industry term with a specific regulatory and safety context.

Understood, but it also get clicks, especially for a company as loathed as Tesla.

Most of the time, these "recalls" are OTA software updates. My guess is that the word "recall" is necessary wording but doesn't imply anymore a mandatory physical fix on the car.

The meaning of the word “recall” has been eroded. At this point we would be better served by renaming it to a more descriptive term.

Dangerous defect notice.

“Initiated safety recalls require a manufacturer's action to announce and remedy the defects.

A recall is issued when a manufacturer or NHTSA determines that a vehicle, equipment, car seat, or tire creates an unreasonable safety risk or fails to meet minimum safety standards.”

So to rewrite the prior example: Tesla issued a dangerous defect notice on all of the vehicles it has sold in the U.S. because some warning lights are too small to meet minimum safety standards.

https://www.nhtsa.gov/recalls

Recall is a term of art in the automotive safety world, as you just posted. It isn’t eroded - if the government decides the fix has to be applied, that’s a recall. Whether or not the fix is OTA doesn’t play into the classification as a recall.

I am aware it is a term of art and that the nature of the fix is independent of the presence of a dangerous defect requiring prompt remediation i.e. a recall.

Unfortunately, lay understanding of the term results in objectively wrong statements like: “it is not a recall, it is just a OTA software update”. We would be better served if NHTSA replaced the term with a descriptive alternative that better communicates the intent and nature of the term.

I suggest: “dangerous defect notice” as it is descriptive, unambiguous, sufficiently alarming, and clearly indicates it is a property of the product instead of the remediation.

I suggest: “dangerous defect notice” as it is descriptive, unambiguous, sufficiently alarming, and clearly indicates it is a property of the product instead of the remediation.

Disagree on this suggestion. “Notice” to me doesn’t imply a necessary response, whereas “recall” does.

But, I do agree with the general sentiment that people don’t understand the legal obligations that come with a recall. And I wouldn’t object to a change of language by the NHSTA, as long as it’s consistent - I would be against any language that tended to differentiate “OTA recall” from “physical recall” for the same reason I don’t like “notice”. That, and the NHSTA may or may not know the full scope of remediation before issuing a recall.

“Mandatory safety update” perhaps?

“Defect” implies a necessary response. A defective product is a product that a manufacturer is legally obligated to repair, replace, or refund. A “defect notice” is thus a notice that the manufacturer is legally obligated to remediate the problem.

It is also consistent with existing terminology, see: “Most decisions to conduct a recall and remedy a safety defect”. We see that the “recall” refers not to the remedy, but to the notice that the manufacturer has obligations to remedy the defect.

Note that “mandatory safety update” is also incorrect because:

“Manufacturers are required to fix the problem by repairing it, replacing it, offering a refund, or in rare cases repurchasing the vehicle.”

So, a update is not necessary, it is usually just the cheapest option which is why it is the most common. Also note how the remediations proposed are consistent with standard defect remediation as I laid out above.

If we renamed regulatory terms of art every time members of the public misinterpreted them, we would be changing terminology so often that we'd end up with everyone being confused, not just some people.

> Recall is a term of art in the automotive safety world, as you just posted.

Also in consumer products in general. Here is a list of recalls from the Consumer Products Safety Commission for example [1].

Sometimes the remedy is to take the product to a dealer to be fixed. Sometimes it is an OTA update. Sometimes it is that the manufacturer will send a repairperson to your home to fix the problem. Sometimes it is a non-OTA update (e.g., there was one for a Samsung washer a while back where if your model didn't have WiFi they would send you a dongle to plug in that would install the update).

[1] https://www.cpsc.gov/Recalls

We use the same term in the med device industry as well. I primarily work on software-as-a-medical-device (SaMD) and a "recall" typically almost always involves a software and/or documentation update.

It’s not warning lights. If you’re getting that wording from NHTSA, they have their heads up their asses.

Especially when the recent recall over driver attention detection was obviously the NHTSA going after Tesla and not a real problem that needed a solution: and the fix seems to have made using autopilot/FSD more dangerous by increasing the amount the screen nags the driver.

Typical low quality journalism. It’s not warning lights, it’s a graphic asset (akin to a CSS sprite if you know web development) and the government wants a few pixels tweaked in an over the air software update to make some icons into i18n-unfriendly English words instead of internationally recognized iconography.

To be fair the subset of icons in question also contained words at a smaller size but this change will lose the iconic look for those elements and substitute an ever so slightly larger font word like BRAKE to be shown in the same placement icon-sized on the screen. The article saying “warning lights” is just par for legacy media these days.

https://static.nhtsa.gov/odi/rcl/2024/RCLRPT-24V051-2023.PDF

> On affected vehicles, the letter font size of the Brake, Park and Antilock brake system (ABS) visual warning indicators is smaller than 3.2 mm (1/8 inch), as prescribed in FMVSS Nos. 105 and 135.

> affected vehicles will receive an over-the-air software remedy

Once every other month I drive about 11hrs two ways. I'll pick a charger every 2-3 hours. 2 hours away the Tesla will say "Preconditioning battery for charging". It's pretty clear this is a bug. It's been there for at least 3 years, as long as I've owned the car. It doesn't give me confidence that FSD is going to work if something so basic fails.

Another thing that happens is driving down an effectively straight highway with 2 lanes on each side and only the cruise control on it will suddenly slam on the breaks in the middle of no where on clear day with no obstructions in sight. Again, no confidence in FSD.

I've also seen your issue where it tells me it steered to miss something but I don't feel any change.

Note: I don't have FSD nor do I have the highway lane change feature.

I have a normal ICE car but it has a supplemental 12v battery for its start-stop system. When that battery gets low/close to EOL, an error message will pop up stating that "steering assist may be limited." I haven't detected any difference in handling however.

Perhaps it is a signal that the 12v battery is not fully charged on Teslas as well.

I've never driven a tesla.

I've also never had my steering wheel randomly stop working in the middle of driving.

I can't help but think reality follows fiction

https://www.snopes.com/fact-check/car-balk/

If GM had developed technology like Microsoft, we would all be driving cars with the following characteristics:

1. For no reason at all, your car would crash twice a day.

2. Every time they repainted the lines on the road, you would have to buy a new car.

3. Occasionally, executing a manoeuver such as a left-turn would cause your car to shut down and refuse to restart, and you would have to reinstall the engine.

4. When your car died on the freeway for no reason, you would just accept this, restart and drive on.

5. Only one person at a time could use the car, unless you bought 'Car95' or 'CarNT', and then added more seats.

6. Apple would make a car powered by the sun, reliable, five times as fast, and twice as easy to drive, but would run on only five per cent of the roads.

7. Oil, water temperature and alternator warning lights would be replaced by a single 'general car default' warning light.

8. New seats would force every-one to have the same size butt.

9. The airbag would say 'Are you sure?' before going off.

10. Occasionally, for no reason, your car would lock you out and refuse to let you in until you simultaneously lifted the door handle, turned the key, and grabbed the radio antenna.

11. GM would require all car buyers to also purchase a deluxe set of road maps from Rand-McNally (a subsidiary of GM), even though they neither need them nor want them. Trying to delete this option would immediately cause the car's performance to diminish by 50 per cent or more. Moreover, GM would become a target for investigation by the Justice Department.

12. Every time GM introduced a new model, car buyers would have to learn how to drive all over again because none of the controls would operate in the same manner as the old car.

13. You would press the 'start' button to shut off the engine.

Power steering failures are relatively common, among companies that have been making cars for a century now.

E.g. https://www.consumerreports.org/car-recalls-defects/gm-recal...

https://www.consumerreports.org/cars/car-recalls-defects/hon...

https://amarolawfirm.com/product-liability/vehicle-recalls/f...

You only hear about this one because it's Tesla.

> You only hear about this one because it's Tesla.

So.. you're saying that power steering failures are common and then proceed to link to articles listing power steering related recalls. But then in the next breath you claim that we only hear about this one because it's Tesla.

Seems a bit contradictory. If we didn't hear about recalls from other car companies, you wouldn't be able to link to examples of them.

Your comment is an example of the "Tesla's just another car company like all the rest" line of defense offered up when there's bad news about Tesla.

> If we didn't hear about recalls from other car companies, you wouldn't be able to link to examples of them.

this is not it lol

> But then in the next breath you claim that we only hear about this one because it's Tesla.

"We" in this context is HN. Type 'recalls' in HN search and see how many Tesla vs non-Tesla recall news were popular on HN.

> Of all the administration's recalls in 2023, Ford issued the most recalls (58 or 16% of all recalls issued), followed by Chrysler with 45, per NHTSA data. That includes either parts of vehicles or entire vehicles recalled, and also compliance concerns over emissions standards or environmental regulations, and software or technology glitches.

> General Motors, Chrysler, BMW and Nissan all ranked in the top 10 most-impacted manufacturers, according to NHTSA.

https://www.usatoday.com/story/money/cars/2024/01/12/vehicle...

How many of the above recalls did we hear about on HN?

Here's one on Hacker News about a Toyota recall from 3 days ago:

https://news.ycombinator.com/item?id=39191931

Maybe look around a bit. Broaden your horizons. Seek out a bigger perspective than only what you see on Hacker News.

You seem to have narrowed yourself much too much.

The only notable thing about the comment section on the Toyota recall, is that it not filled with apologists crying that people are unfairly highlighting a safety issue.

Of course, Toyota is treated as a real car maker, and they should be scrutinized as such. Recalls are no joke.

But when it's a Tesla recall, the comment section is a shitshow. An odd mix of Musk sycophants, shills that are financially invested in the company, and fanboys that adopted Teslas as a sort of status symbol.

The funny thing is that initially I was neutral on Tesla. I didn't love or hate Elon, and saw Tesla as another US car maker, not more or less special than Ford, GM, etc.

But those people really soured me. If the consumers of those cars are willing to undermine its flaws for ulterior motives, that's one car I'll always refuse to buy.

[flagged]

So.. you're suffering from a persecution complex about a car brand?

I think there are bigger and better things to worry about than Tesla's image. You're not even getting paid to be Tesla's volunteer PR flack and spin doctor.

If you want a career in PR then at least get paid for it.

Just correcting people and noticing things is a persecution complex now? And nothing about all the haters of a car brand? Literally I just posted a list of URLs and stated that other car companies have had power steering problems too and it triggered a bunch of angry replies and got downvoted for listing inconvenient facts.

You're not just correcting people. You're saying things like "irrational Tesla haters" and "upvoters bring almost every negative Tesla story and hit piece on to the front page" and "you only hear about this one because it's Tesla".

That's less "noticing things" and more paranoia. And paranoia about something as trivial as a car brand. That is a significant loss of perspective.

An obvious problem with Tesla is its lack of commitment to the truth. When lying becomes standard operating procedure for a company then that company is not worth your time or money.

In 2016 Tesla claimed that "as of today, all Tesla vehicles produced in our factory – including Model 3 – will have the hardware needed for full self-driving capability at a safety level substantially greater than that of a human driver". That was lie: https://www.tesla.com/blog/all-tesla-cars-being-produced-now...

Tesla's video "demonstrating" full self-driving was staged: https://www.reuters.com/technology/tesla-video-promoting-sel...

Musk's cavalcade of lies about full self-driving, year after year: https://motherfrunker.ca/fsd/

Tesla even feels the need to lie about something as dumb as quarter mile times for the Cybertruck. Their recent "quarter mile" video was fake. Lying is simply the company culture at Tesla:

https://insideevs.com/news/699260/tesla-cybertruck-porsche-r...

https://www.youtube.com/watch?v=HRYS5VWXZts

So instead of running around with paranoid fantasies that world is out to get Tesla, first ask yourself why Tesla feels the need to lie so often. Then ask yourself why it is that you have to spend so much time and effort making excuses for Tesla.

Maybe it isn't the rest of the world that's the problem. Maybe all those articles that you don't like aren't "hit pieces" after all.

Maybe the problem is Tesla.

I could make a long list of 'lies' that other car companies made up with actual harms that are completely ignored and never brought up. Like the Volkswagen emissions scandal or the GM ignition scandal that resulted in deaths and criminal charges for hiding and falsifying information to the govt.

https://www.washingtonpost.com/news/business/wp/2015/09/17/w...

https://en.wikipedia.org/wiki/General_Motors_ignition_switch...

Even if what you said was true it's not okay to constantly make up misinformation about Tesla. As you listed there's enough to legitimately criticize Tesla and other automakers about, so why are people hell bent on making up things?

> So instead of running around with paranoid fantasies that world is out to get Tesla

There's a long line of shortsellers, fossil fuel and car company anti-EV lobby money in the media and just plain haters making up things, and anyone pointing out the falsehoods personally attacked.

I hate this trend of demonization of cancelling something or someone, and then falsehoods are spread and anyone pointing out the falsehoods are demonized.

> I could make a long list of 'lies' that other car companies made up with actual harms that are completely ignored and never brought up.

Ah, back to the "Tesla is just another car company like all the rest" whataboutism.

> Even if what you said was true

It is.

> it's not okay to constantly make up misinformation about Tesla

So.. it's made up misinformation that the National Highway Traffic Safety Administration is doing an engineering analysis of Tesla's power steering problems?

> There's a long line of shortsellers, fossil fuel and car company anti-EV lobby money in the media and just plain haters making up things, and anyone pointing out the falsehoods personally attacked.

There's that paranoia again.

> It is.

Not that's your paranoia. Like how people on here said Cybertruck was scan for the preorder fees and would never ship based on similar thoughts like yours.

> So.. it's made up misinformation that the National Highway Traffic Safety Administration is doing an engineering analysis of Tesla's power steering problems?

When did I ever say that? I just noted that most other automakers' having similar issues and recalls don't reach the front page of HN. Ford had 68 recalls in 2023, how many reached HN's front page compared to Tesla's?

I was responding to a comment high up the page stating this:

> I've never driven a tesla.

>I've also never had my steering wheel randomly stop working in the middle of driving.

The implication is that only Teslas have power steering loss issues because the poster never had a Tesla and never had power steering losses, based on a faulty anectdotal assumption. Isn't that misleading if not misinformation?

Mazda just had a big recall for power steering loss:

> Mazda recall for power steering warning Mazda is recalling 43,752 of its 2024 CX-90 vehicles. An assembly issue with the worn gear in the steering wheel system could result in a sudden loss of power steering assist. Loss of the power steering assists while driving increases the risk of a crash, the NHSTA reports.

> Mazda owners can visit their dealers to replace the spring engaging in the worn gear. The mechanics will reapply grease to the gear teeth, which will be free of charge. All Mazda owners affected will receive a notification letter on March 18. Owners can contact the Mazda customer service at 1-800-222-5500 and press option six. Mazda's number for this recall is 6542A.

And this isn't even a software recall that is fixed with an OTA update.

Where's the coverage of that story on HN?

I responded with a list of URLs showing other car companies had similar issues and somehow everyone's super angry about me and it got downvoted as punishment for adding context and facts, and to bury it lower down the page. And all the personal attacks.

> Not that's your paranoia.

So.. your claim is that Tesla didn't lie about its Cybertruck quarter mile video? After the lie was exposed Tesla admitted that they didn't complete a quarter mile.

At this point you're not even agreeing with Tesla. These are the dead end roads paranoia leads you down.

You're just putting words in my mouth now because you are completely unable to address the arguments in the rest of my post because you have to admit they're correct. You also admitted other car companies lie by saying I was resorting to whataboutism.

It's not whataboutism when someone implies power steering failures happen only with a particular car company, and someone corrects them by showing instances with other car companies. That's irrationalism.

You're not even getting paid to be anti-EV and fossil fuel lobby's volunteer PR flack and spin doctor to spend all that time and effort to demonise Tesla while completely ignoring the other car companys' faults.

If you want a career in PR then at least get paid for it. Literally promoting Volkswagen all over the place after the intentional emissions scandal ruining people's health with 5 million dying because of pollution every year, that's sad, disgusting and pathetic.

> You're just putting words in my mouth now

I sure hope not. Your words aren't consistent and don't make a lot of sense. I like to think more clearly than that.

> It's not whataboutism

I don't think you've fully understood what whataboutism is. Here's a helpful link:

https://en.wikipedia.org/wiki/Whataboutism

> If you want a career in PR then at least get paid for it.

Ah, so really you're the one putting my words in your mouth.

You seem to be very obsessed about a car brand and building an identity around hating it and spreading the hate around as much as you can for more than half a decade.

I think there are bigger and better things to worry about than Tesla's image.

No, I'm just objective. You perceive objectivity as hate because you've lost perspective.

Loss of perspective is a problem common to all cults.

If you were anywhere near objective you would give the same scrutiny that you give Tesla to other car makers. Since you're horribly biased against Tesla you instead try to make them look good in your comments and submissions. How is that being objective???

You really haven't understood this whataboutism thing, have you.

Your paranoia is making you foolish. You really do need to sort it out.

Flamewar like this is completely unacceptable:

https://news.ycombinator.com/item?id=39248250

https://news.ycombinator.com/item?id=39246730

https://news.ycombinator.com/item?id=39246626

https://news.ycombinator.com/item?id=39246316

https://news.ycombinator.com/item?id=39245999

https://news.ycombinator.com/item?id=39245589

https://news.ycombinator.com/item?id=39244494

https://news.ycombinator.com/item?id=39237192

https://news.ycombinator.com/item?id=39236997

Since you have a long history of posting like this, I've banned the account.

If you don't want to be banned, you're welcome to email hn@ycombinator.com and give us reason to believe that you'll follow the rules in the future. They're here: https://news.ycombinator.com/newsguidelines.html.

Flamewar like this is completely unacceptable:

https://news.ycombinator.com/item?id=39248035

https://news.ycombinator.com/item?id=39246692

https://news.ycombinator.com/item?id=39246469

https://news.ycombinator.com/item?id=39246170

https://news.ycombinator.com/item?id=39245886

https://news.ycombinator.com/item?id=39245065

https://news.ycombinator.com/item?id=39241621

https://news.ycombinator.com/item?id=39241564

https://news.ycombinator.com/item?id=39237959

https://news.ycombinator.com/item?id=39237162

Since you have such a long history of posting like this, I've banned the account.

Please don't create accounts to break HN's rules with.

I see they've mastered the modern rhetorical technique of having a trillion in assets, billions in sales of this product, buildings filled with professional engineers, lots of bragging about the size of their engineering genitalia ... and instantly adopting the position of the fragile, vulnerable victim the moment they are criticized. They are so agile. It's like criticizing kindergartners - they start crying and then what do you say (I don't criticize real kindergartners, only Tesla engineers and executives). I'll also point out that Musk delights in demonstrating that he doesn't give a crap about anyone or anything, but let's not stoop to that level.

> Power steering failures are relatively common

Let's have a contest: who can describe the problem most euphemistically? 'Minor safety issues are common'? 'Software calibration issue'? 'All software has bugs'?

> I don't criticize real kindergartners, only Tesla engineers and executives

Where is this criticism? And why only Tesla engineers and not other car company engineers or executives? Why are they exempt from criticism?

right, it's a software bug, not a physical problem. That's where the absurdity lies.

good evening tesla apologist, look closer. Those are due to physical problems, not software bugs.

As most of the apologists in this thread have rightly pointed out, Tesla can fix the problem with an over-the-air update. Other manufacturers can't.

If you took my post to be claiming other cars never had physical problems, you misunderstood the absurdity.

We hear about this one because it's fucking absurd for a software bug to cause intermittent steering issues.

and tesla's aren't that old, give it 20 years and then you might have a leg to stand on comparing recall percentages. Toyota just recalled 2003/2004 corrolla's for an airbag issue. Tesla hasn't had time to find all the bullshit yet.

> As most of the apologists in this thread have rightly pointed out, Tesla can fix the problem with an over-the-air update.

Huh? That's for a different issue with the UI icons being small. Source?

> We hear about this one because it's fucking absurd for a software bug to cause intermittent steering issues

Source that it's a software bug? In fact the article says it's likely a physical defect in parts.

> good evening tesla apologist

Ah, irrational Tesla haters are not known for the ability to discern facts.

I don't always have physical defects, but when I do, I fix them with non-physical software updates.

> 13. You would press the 'start' button to shut off the engine.

Is this not how pretty much all modern cars work?(that use push-button ignition, which I think is almost all of them?)

It’s how my 2010 Toyota works - if the engine is off, and I press the button (while foot is on the brake) the engine starts. If I push the button while the engine is running, it turns off. This seems like sensible behavior to me

It's a joke, but modern SV has embraced the absurdity. I suppose Microsoft embraced it back then (the 1990s?), in a way, but I guess I assume that the software industry was just immature back then and hadn't developed effective engineering.

My coworker experienced #4 with his Teslas . Just randomly shut down while driving had to be rebooted

An engineer (for whom I do occasional handyman work) recently disclosed to OEM a "pending public safety notice" that Tesla had an egregious craftsmanship issue within the chasis of a major product.

Tesla's initial response was "can you please wait longer to disclose this?"

Just another Boeing-like example of why industry self-regulation/compliance doesn't really mean shit when profit-seeking for share-holders is legally directive #1.

As Tom Rowe (and so many others) brilliantly states, "Safety THIRD" ...

"An OTA update will avoid this" "When it can be updated it is not a recall"

It will always be registered as a recall if safety is involved, even if it is just an OTA. Registering it as a recall is a legal requirement.

"It still counts as a recall"

It literally says in the very next sentence: "Documents posted Friday by U.S. safety regulators say the recall will be done with an online software update."

You cutting that out has to be intentional. ;)

I'm kidding: I don't think you or the writer had any intent of anything, other than communicating, nor do I think about that sort of thing.

Hopefully if there's some actual problem with the steering, they'll get a break from all the politically motivated government persecution long enough to fix it.

Reply on news.ycombinator.com