Hacker News new | past | comments | ask | show | jobs | submit login
Electric cars prove we need to rethink brake lights (youtube.com)
161 points by trymas 3 days ago | hide | past | favorite | 209 comments





Just watched this while eating. It's half an hour long, so here's a summary:

- He owns a 2022 Hyundai ioniq 5, USA version

- The brake lights generally work fine, however when in one-pedal mode (i.e. accelerator does regenerative braking for the first X% of travel) they only activate when the vehicle is about to stop so long as the driver has the accelerator ever slightly depressed

- This is a problem because the regenerative braking is quite strong - he goes from 60 mph/96 kph to 0 in 13 seconds and the stop lights never illuminate

- He thinks this should not be allowed and Hyundai should recall the cars to fix this

- However, in the US this is perfectly legal - regenerative braking is not considered as part of the "service brake", so illuminating the stop lights is not required

- The EU on the other hand just fixed this gap in the law in March, by requiring that stop lights be illuminated whenever the deceleration exceeds 1.3 m/s^2


I have actually thought about this when driving a Tesla which also has the regenerative braking. Here's what their website says for the record, it sounds like it works exactly how I would want, just depends on the threshold for deceleration they set:

""" During regenerative braking, Tesla will still activate the brake lights when the vehicle is slowing down, even if the brakes aren't being used at all. Tesla determines whether to turn on your brake lights based on your vehicle's rate of deceleration. If you're unsure if your brake lights are on, look at your Tesla screen, the car in the display shows the brake lights lit up when the brake lights are activated. [0] """

[0] https://www.notateslaapp.com/tesla-reference/1051/how-tesla-....


And this is where I take a minor quabble with TC. He thought that GM “over engineered” the volt’s brake light system because it used an accelerometer. But honestly I think it is the easiest way. The only “trick” is to make it stay on once stopped (I’m sure there are conditions when you wouldn’t want them on when stopped though).

In fact it looks like the new standards are going to be defined by the rate if deceleration regardless of the means used to do so. Which means basing the lights on an accelerometer is the most reliable way to meet the spec. It shouldn’t matter how you are slowing down, just that you are.


Why do we need an accelerator when we know how fast the wheels are turning directly? The first derivative of wheel speed is almost always car acceleration. Is this to handle some edge cases like not turning the brake lights on when we're actually sliding on ice with the wheels locked? Seems unnecessary.

Why is it unnecessary? Isn't the primary goal of brake lights to inform people of deceleration and the secondary goal to indicate when a vehicle is locked stationary?

Showing a speeding vehicle behind you that you've lost control and should slow down is well within the conops of brake lights.


I think there's an agreement on the goal.

I believe the poster is asking why do we need another sensor to achieve that goal? We already have a speedometer based on wheel motion. A trivial computation gives acceleration.


Because tires can block, not all tire turn at the same speed all the time, because redundancy is a good thing... There a lot of reasons. And if the number of sensors in your car bother you, well, the early 80s, with carbs and without ABS, are last model years you can buy.

No need to get snarky :)

Redundancy is a good thing. Do we feel manufacturers are cross-checking accelerometer with the speedometer? If not, there is no redundancy gained.

On the other hand, if your speedometer is faulty, or not working, you'll know pretty fast. If your dedicated accelerometer is faulty, you might have no idea what your brake lights are doing in the back.

Dunno; it just feels "we need to know acceleration, let's add an accelerometer" is a non-imaginative, add-cost, add-complexity idea. Again, if we think ABS and speedometer and the new accelerometer are being cross-checked and sanity-checked, awesome, but I'm just a bit cynical of that, compared to adding the 3rd thing to do the same thing.


> add-complexity idea

You're on a forum where a lot of people think Kubernetes is a good idea.

;-)


No idea what Kubernetes are, and I am serious here, since I am no software dev. On the hardware side of things, there is redundancy and complexity. Those two might look similar on the surface, they are totally different beasts so.

To simplify it as much as possible, Kubernetes is a system that deploys and manages containers. Containers allow multiple programs to run under the same system while being completely isolated.

However, Kubernetes is actually incredibly complex. It's powerful, but holy hell does it add complexity. And yet it's basically a buzzword and people are using it in very basic scenarios that don't call for the complexity that Kubernetes adds.

So that's where my joke comes from. There are people that behave as if adding unnecessary complexity is a feature.


Thank you a lot for the explanation!

Many late 80s/very early 90s fuel injection vehicles have shockingly low numbers of electronic sensors also. MAP, O2, TPS, CPS is all you really need. Some of them are even analog.

An accelerometer will show braking when going at constant speed down a hill. Change in wheel speed is better.

I’d say you are braking somehow if you are going at a constant speed downhill, are you not?

What do you save by not having that sensor? Nothing. What do you lose by not having that sensor? The ability for the system to perform its requirements.

I'm assuming you read the context of loss of traction being a requirement.


> What do you save by not having that sensor?

A $9000 repair bill in 6 years when your mechanic tells you “Sorry your car is immobilized and won’t start, the deceleration brake light sensor went and we need to remove the motor and 3/4 of the wiring harness to replace it”. And for anyone who thinks I’m being sarcastic, try owning a BMW or an Audi and you’ll know it’s the truth.


I have owned a troublesome 2010 BMW for 8 years now. No individual repair has been over $2,700.00. Most are closer to $1,500. I was quoted (by a dealer) $15,000.00 for engine repair once but it turned out to be a spark plug. $9,000.00 sounds like a dealer quote. Find a good independent shop.

A MEMS accelerometer is pennies. If it is critical then add redundancy and don't accept single fault tolerance. If it is optional or the signal can be estimated from other sources with acceptable error then fail gracefully. This is honestly simple stuff. Accepting less from manufacturers is a bad deal.

>>A MEMS accelerometer is pennies.

That's an dishonest argument (perhaps not intentionally). NOTHING is "pennies" to a consumer when it comes to repairing a vehicle. More frequently, you pay $1,500 in labour and parts to replace something that costs "pennies" in some bulk manufacturer wholesale catalogue.

Modern cars are getting more and more awesome, in terms of safety and convenience; but the sticker shock when going to dealership for repair is also becoming bigger and bigger, and it absolutely is intertwined with the significant effort to make 3rd party or even self-repair difficult to impossible.

So again, IFF this is an easily replaceable part that is thought-through and cross-checked intelligently with other existing sensors, brilliant. But can you at any level understand my skepticism that any of these are true? :)


I'm not talking about the price of maintenance but the cost of manufacture. Please give generous interpretations rather than ones that fit an argumentative narrative. Assuming the topic is cost of repair rather than manufacture makes no sense when we're talking about system engineering to accomplish functional requirements.

The problem there is owning an Audi or BMW, not the sensor itself.

Relatedly, having a dedicated accelerometer sensor is a baseline dependency of many of other modern features like cross-comparing compass headings and GPS for map heading information (much less any of the Level 2+ "self-driving" mechanics). Most cars likely want one, anyway, even in base models. It's one of the cheapest sensors in a suite of increasingly standard sensors (in almost any form factor of device, not just cars, but phones/watches/toasters/etc).

One more possible reason might be when driving down a hill: When descending a sufficiently steep incline, I may be braking just to maintain my speed! That said, I'm still _feeling_ negative acceleration to maintain a constant speed. (That said, if I'm doing this for more than a handful of seconds, I'll generally downshift, and _that_ doesn't trigger brake lights, despite also causing increased negative acceleration. So maybe regenerative braking is analagous, and fine without brake lights?)

> Why do we need an accelerator when we know how fast the wheels are turning directly?

Because the car doesn't precisely know the wheel's radius. The radius depends on the rim, tire, and tire pressure - and the car's operator may have accidentally or deliberately chosen something unexpected for any one of those 3 parameters.


You're not wrong but how much does not precisely knowing the wheel's radius change things?

The operator would also experience an incorrect speedometer. For the purposes of a brake light it'd be off (either lighting up too soon or too late) by some amount but I imagine it'd be "close enough" except for the most pathological cases of tire sizes.


You don't need to know the wheel radius, you just need to the rate of change of the wheels RPM. Then the brake light could be set to some conservative value where regardless of any realistic tire size/inflation scenario the brakes will activate at a reasonable deceleration.

In any case, with standard gas/brake pedals on ICEs, I can barely tap the brakes and have my brake lights turn on without any deceleration on my part, or I can do that and still be pressing the gas and so actually be accelerating with my brake lights on.


This is not a problem. We know this isn't a problem, because ABS almost always uses wheel speed sensors to do it's thing, and ESC usually taps into the same wheel speed sensors. Hell, those sensors are often used to tell you when your tires are low on pressure, and yet that doesn't prevent any of the other functions from working, or even affect your speedometer enough to care.

Changing wheel size enough to affect readings of speed does not change the broad slope of the derivative of wheel speed.


ABS relies on the fact that the computer knows how fast the wheels are turning in relation to each other. Wouldn't seem to be much of a leap to leverage that system to determine if the car is slowing down.

If the car doesn't know the size and speed of the tires it also can't accurately display the current speed, which seems like at least as pressing a problem.

Edit: Or is detecting deceleration more sensitive? I guess I don't know how precise it needs to be relative to how precise a speedometer needs to be.


I don’t think it’s such an edge case at all. I imagine it would be disconcerting to the driver behind you if you slammed on the brakes on a wet road, locked the wheels into a skid, and the brake lights went out. (Shouldn’t happen with modern braking systems, but still…)

Could be as simple as "the controller doesn't have wheel speed on its can bus"

>The first derivative of wheel speed is *almost* always car acceleration.

There you go, you answered your own question. When it comes to safety features, "almost" always working isn't really good enough.


Yea the Tesla brakes work exactly how you would want. As soon as you start decelerating, even w/ regen, the lights kick in. Not sure how Hyundai messed it up after having an example..

Hyundai/Kia will do the minimum necessary to comply with the local laws.

Interesting, would seem more efficient to comply with as many laws as possible that don't clash and have one configuration, only tweaking for local oddities.

I have a complaint about Tesla on this one. In Europe they ship vastly better tail lights with a different color for turn signals. As I understand, the superior signals are acceptable, though not preferred, by US standards. But Tesla makes a separate, different, worse part to use in the US, rather than giving us the good stuff also and reducing their manufacturing variation.

As an European I always wondered about why in the US they have red turn signals. It seems...odd, so to speak?

It’s cheaper (or was) - early cars had one red light on each side and could blink it for a turn signal, but it was also the brake.

Later a second light was added for the taillight but they did it in the same bulb.


Thanks for the explanation! It still seems odd to me that it hasn't been regulated somehow, having 2 different colors for 2 different actions is safer.

Suburban and Americans aren't so good at living with the level of aggression that city driving requires, and so prefer having clearly delineated left turn time to something perceived as fighting in traffic.

It's a pattern for Hyundai/Kia. The whole thing with "no anti-theft immobilizers" is only where it's not mandated.

Their business is shaving pennies, hence why they didn't install immobilizers in states that didn't require them, in order to save a few bucks per car.

My 2020 Bolt does the right thing also. I only wish the lights would stay on while I'm stopped at a light but that's minor.

Do you not hold the brake when you are stopped? I know electric cars don't have to creep like an automatic ICE, but I drive stick and still hold the brake at a stop because not randomly drifting forward or backward is safer and it also is important in the rare case of getting rear ended

I don't know about GM, but in a Tesla, if you have 1-pedal driving enabled, then it automatically holds the brake. And the brake lights stay on.

> I know electric cars don't have to creep like an automatic ICE

They don't have to, but funny enough, it's an option in a Tesla. My wife uses it simply because it's what she's used to.


One drawback of this is the person behind may incorrectly assume you’re brake checking them; i have had a number of tailgating drivers retaliate by accelerating in front of me and slamming on their brakes (which is especially stressful with a kid in the car). Stuff like this makes me deeply regret choosing Tesla.

I would love a control panel option to turn automatic brake lights on/off.

It's really annoying to signal something to others that you don't intend.

I think there are a bunch of things tesla does that prevent you from being a good driver.

some examples:

- you can't flash your headlights (unless you traverse menus on the touchscreen)

- you can't set the wiper interval

- quickly turn on the defrost?

- turn on the hazard lights? (you should google it, it's funny)

newer cars go even further - no stalks, touch controls for turn signals, horn and wipers on the steering wheel


> - you can't flash your headlights

You can flash the high beams which is good enough, IMO.

> - you can't set the wiper interval

You absolutely can. In my Model 3, if I press the button on the left stalk, it makes the wipers wipe once and shows the wiper options on the screen. I can set them to Off, Slow, Medium, Fast, and Auto.

The only exception is when the car is on Autopilot. Then, it's forced to be Auto. The quality of Tesla's auto-wiper software is questionable, though.

> - quickly turn on the defrost?

There's a button for that now.

> - turn on the hazard lights? (you should google it, it's funny)

? It's a button above the mirror?


> ? It's a button above the mirror?

Yep. FWIW, Hazard Lights is the one control that is mandated to be a physical button by US law.


> You can flash the high beams which is good enough, IMO.

flashing high beams at night is a good way to set people off and wind up in a road rage incident. some people interpret that signal very aggressively and will retaliate.


What model and year do you have?

My 6 month old Model Y has a dedicated button for hazard lights on the ceiling.

I can flash lights by pulling the left stalk (I forget if forward or backwards).

Wiper interval I do miss but what I usually do is either hit the wiper button once on the stalk and set via screen or use the voice interface to say “wipers speed 1”.

Can’t you turn defroster on with the voice interface too?


I feel like the psychology behind being a driver is a problem.

The BMW i3 is the same way.

This sounds like a recallable defect in the OP’s car’s software.


The Mustang Mach E is somewhat similar, if I remember correctly.

When slowing down with regenerative breaking it triggers the lights based on how fast it’s slowing down.

So if it’s just a tiny bit (slowing from 35 to 30 over a few seconds) it won’t kick on, but from 70 to 35 over a few seconds would.


We had manual transmission cars that could slow down from engine braking for decades and that never lit the brake lights. I'm not convinced this is an issue. It is your responsibility to keep a safe following distance from any car in front of you regardless of circumstances. You should be able to safely not rear end the car in front of you even if it immediately hits a brick wall and stops completely, and even if the brake lights are entirely missing.

> You should be able to safely not rear end the car in front of you even if it immediately hits a brick wall and stops completely, […]

I don't think that's the usual expectation. Maintaining absolute braking distance is what railways do, whereas cars commonly operate on relative braking distance plus a safety margin accounting for your reaction time.

Over here, both driving schools and other general road safety education talk about keeping 2 seconds following distance (or half your kph-speed in metres, which corresponds to 1.8 seconds following distance), and traffic engineers commonly give the capacity of a single lane of traffic as 1800 vehicles/hour, which again corresponds to 2 seconds distance (this time as measured vehicle front to vehicle front, instead of tail to front of following vehicle).


Yep.

I drive a Jeep Wrangler and between its manual transmission and the fact that it's less aerodynamic than a cow, I can and often do slow from highway speeds to ~25 MPH without touching the brakes. It's never been an issue.


EU regulators were convinced it was an issue. And do many drivers follow you so distantly?

> he goes from 60 mph/96 kph to 0 in 13 seconds and the stop lights never illuminate

I ride a motorcycle and I give Teslas a lot of cushion when riding behind. They're common enough in the bay area to be treated specially, because of the one pedal driving (my car is set to "creep", but it is still strong at 70mph).

The way the car changes speed and the way the Model S looks from behind, it takes a bit of time before you realize it is getting too big in your PoV without lights coming up, particularly as people lift off through turns or highway onramps.

The cars which are bigger or higher up like an ID4 or Rivian seem to not suffer from this sort of "hidden width" that the Tesla has, the wheel flares melt into the backdrop when looking at it.

I don't think Technology Connections is ever going to drive a Tesla or test it, but the low profile fast cars are much harder to figure out speeds off from parallax without the brake lights.


Engine breaking on a motorcycle can be quite significant as well. I've thought about this being a problem there too.

In the safety classes I’ve taken, there’s usually been training (albeit brief) of being aware if others are behind you, but building the habit of using the engine to decelerate that you flash your brake lights and or apply enough brake to trigger them without heavily braking.

This was also an issue when I had a larger manual car with a medium size (4.6l) v8.

At lower speeds <45mph, it was entirely possible for me to be significantly decelerating using only the engine, where I would make sure to at least flash my brake lights several times.


Diesels with a J-brake can also decelerate quite abruptly, though in that case there's still a remarkably loud audio queue from the engine compressing

I own a Canadian Ioniq 5 and it is indeed quite dangerous. I have been honked a few times in my year of ownership. Now, I usually release the pedal completely to at least flash the brake light to show people behind me that I am decelerating. But my wife hate the sensation so I usually don't use 1-pedal when I am with her. Which is sad because I really enjoy driving in this mode.

I totally agree, when you are decelerating at more than x/(m|k)ph light should come on.

On a different matter, first generation car, and there is a huge flaw with a high percentage of Ioniq5/Ev6/Genesis GV, have a ICCU (Integrated Charging Control Unit) failure at around a year of ownership. My car at the dealer for over a month without an ETA on part delivery :(


In manual cars, it's always felt wrong how much and how fast the car can decelerate by changing to lower gears without the brake light turning on - even if you are not technically using the brake, you can lower the speed so much as to create a dangerous situation with the guy coming behind, and therefore turning on the brake light would be appropriate.

It’s a feature not a bug. Everyone immediately SLAMS the brake when they see a cop and it’s a dead giveaway if you want to get a speeding ticket. Engine braking has saved me more than once because you can slow down without being an immediate target for the cop.

Exhaust brakes on truck or engine braking behaves the same way. There can be strong deceleration without brake lights.

Deceleration threshold is great solution because it signal what is happening.


Trucks use exhaust braking because it's quicker than braking using the brakes without overheating. They can get away without brake lights only because the noise makes it obvious when trucks are exhaust or engine braking.

Engine breaking, which any manually car driver uses, does not produce any indication that you're slowing down. Technically, the engine noise is louder, but not enough to notice outside of the car.

You're right of course, but these days engine braking is most commonly used by tow vehicles, whether that's a family car or a big hauler. Either way when towing, these are generally required to have trailer brake controllers that are activated by deceleration so that the trailer brake lights do come on.

When we cross the Cascades or Rockies in our Land Rover Discovery towing an Airstream, we have to shift and engine brake on the downslope for miles (to avoid overheating the breaks) and it's noticeable that the trailer brake lights go red, while the Land Rover doesn't unless I tap the brake pedal.


You can engine brake in any "manumatic" like Subaru with their flappy paddle CVT. You could drive a manual without engine braking more than a normal automatic; nothing forces you to downshift. You could also only up shift and use neutral when braking with actual brakes, which I think a lot of people do when learning manual.

Indications that the car in front of you is breaking include brake lights, distance between you and them, and weight transfer to the front (nose down, rear up). Only one of those is affected by the thick pedal.


It's illegal to drive in neutral. And of course you can engine break in automatic cars, but you're not likely to.

>It's illegal to drive in neutral

What law says that?


Dunno about other states, but in Oregon, it is explicitly forbidden to coast in neutral downhill

https://oregon.public.law/statutes/ors_811.495

> A person commits the offense of unlawful coasting on a downgrade if the person is the driver of a vehicle on a downgrade and the person coasts with the gears or transmission of the motor vehicle in neutral or with the clutch disengaged.

However, this only refers to going downhill. It says nothing about approaching a red light or any other scenario where you might consider choosing to coast.


Thank you for the summary. I understand how creators are incentivized by YouTube to produce longer content, but it is a bit ridiculous that this 30 minute video could have been 6 tweets.

Of all the YouTubers out there, I wouldn't call out Technology Connections for padding their content. Sure, you can get a short summary of the useful information, but the show is about more than "short collection or facts". It's like saying "it's ridiculous that an episode of Mythbusters is half an hour, you can fit the results in a tweet!"

To go off on a tangent, someone went and edited Mythbusters episodes to remove the fluff and called it Smyths. There's a torrent of it and it's a way better watch.

I've seen those, and while I understand it, I feel it loses a hell of a lot of the pathos of what mythbusters was. A huge amount of the goodness is the interplay and personality of Adam and Jamie, and it gets lost I think in those edits. Add that to Mythbusters not exactly being scientifically sound, and more like "these TV effects artists build a weird contraption that MAYBE answers a question that isn't well formed". Cutting out the recaps is useful though.

I give Mythbusters a pass because the padding is intentional entertainment (that I happen to enjoy, obviously).

I don't watch TC because the padding doesn't come across as entertaining to me; it feels like a lot more beating around the bush. There's a _lot_ of repetition.


I would. He repeated his points several times.

I absolutely would say that about Mythbusters.

Some of us still have semi-intact attention spans. TC caters to that kind of audience.

Imagine how much you could brag about your attention span if the video was 8 hours long!

If you haven't watched a 4 hour youtube video essay by one of the greats, you are genuinely missing out.

I watch long videos at any time. Usually during coding. For instance this summary and review of utopia: https://www.youtube.com/watch?v=PFx2QM0Z8Qo 7 hours : 41 minutes, watched it one go.

I love AltShiftX book and movie reviews. The Dune ones are long and interesting.

> but it is a bit ridiculous that this 30 minute video could have been 6 tweets.

In the video, he researches relevant laws in the EU and US, and explains the details and implications of those laws. He compares to previous solutions from another manufacturer, GM, and also personally verifies that his car's automatic cruise control mode does properly use brake lights, and also shows how that is following the letter of the law.

There is no cruft in the video, just a lot of densely packed content. Yes you can summarize the video in a few sentences, but you miss out on a lot of the details.


I watched the video for like 4 minutes and it was 4 minutes of him saying “the brake light doesn’t come on until your foot is all the way off the pedal.”

Do you watch sporting events and say, “why did this take so long? They could have put it into one tweet: the Knicks lost again.”

> Thank you for the summary. I understand how creators are incentivized by YouTube to produce longer content, but it is a bit ridiculous that this 30 minute video could have been 6 tweets.

This isn't even the worst example, because the additional content is at least part of the entertainment (even if not the most efficient).

Padding is par for the course among Youtubers. If you play a lot of video games, you'll notice this, because gaming news and tutorials are often done on Youtube.

There are a couple of different thresholds for algorithmic promotion and monetization - I believe 10 minutes is the biggest one - so a lot of videos are padded with filler content just to make the threshold.

You'll also see it on videos for "how to pronounce ______", because they can't monetize below a certain threshold, so they'll pad it with useless filler in order to make the video over 30s long.


TikTok really is killing online video...

Imagine if Vine had succeeded.

Imagine if Elon brings back Vine....

tbh, I liked Vine. It allowed someone to write a joke and punchline without allowing them to fill it in with bullshit.

How much money would they have made from those six tweets?

Content follows the money. Even if it’s anemic like YT, it’s still a lot more than zero.


Our 2014 BMW i3 lit the brake lights when regenerating...hard to say how firmware changes may have altered this behavior over time, and there were firmware changes for our US based i3, but the Ioniq situation in the post seems like an expensive accident waiting to happen.

> they only activate when the vehicle is about to stop so long as the driver has the accelerator ever slightly depressed

Do the brake lights come on if you don't have the accelerator depressed at all?


Yes they do, but releasing it all the way is jerky. To fix this, EVs should limit the jerk rate rather than introducing the brake at full power when you release the accelerator. This is how it’s done by most metro trains.

Note: one-pedal mode slows the car as the accelerator is released.

> - He thinks this should not be allowed and Hyundai should recall the cars to fix this

Why do a recall? Sounds like it's a simple software fix. Can be deployed over the air.


Software fixed are "recalls"

Yup.

Which is why it's so annoying when Tesla recalls hit the news. So many people treat it like a huge event, when really, it's just a small software update.


AFAIK eGMP (platform for Ioniq, EV6, GV60) OTA updates are limited to infotainment system - you have to go into the dealership for major system updates. (I have an EV6)

My 2021 Hyundai Kona EV illuminates the brake lights immediately when the regen kicks in. Strange that that's not the default behaviour on the other model.

It's quite annoying that Hyundai got this so wrong, while a long list of other makers got it right. Even without regulators demanding they do so.

And there I was naively thinking the new EU rule existed, and was applied by OEMs, since regenerative breaking is a thing...

We should determine what rate of deceleration counts as 'coasting', and anything beyond that counts as braking.

Unless said legislation/determination only applies to electric cars, they'll have to give manual ICE vehicles a look-at as well; my TDI for instance can engine brake so well that it behaves very similarly to regenerative braking, and the service brakes get used only where absolutely necessary. Transport trucks with exhaust or Jake brakes and a light(er) loads can get often get away with similar-to-EV single-pedal driving regardless of whether the transmission is automated or not.

This might be a hot take, but if you're actually paying attention to driving it's not exactly difficult to see a vehicle decelerating in front of you, brake lights or not.


I'm only on my second set of brake pads on my TDI and you're right. However in most real driving situations, engaging the engine braking requires feathering the brake long enough to get it to downshift.

So while my brakes aren't on the whole time, they are on when I first make myself a problem that needs to be managed.


>engaging the engine braking requires feathering the brake long enough to get it to downshift.

What? No, you put in the clutch, pull out of gear into neutral, rev match up to the engine speed for the lower gear, plop the shifter into it, release the clutch. What do you use the friction brakes for? Hell, you can skip the rev matching completely if you don't mind clutch wear and jerkiness.


Even with a DSG-equipped car like the parent likely has, you can use the "manual" mode ("M" on the shifter) and downshift to as low as the car will allow. This also has the benefit of 'preparing' the transmission for further downshifting, as it will pre-select the next gear down instead of up. My car is indeed a 6MT, and downshifting happens exactly as described.

I believe they're talking about a manual transmission TDI. In which you could simply downshift and engine brake, without engaging the brakes.

The EU has defined it as a deceleration of 1.3 m/s^2.

This isn’t really a new problem. You could engine brake your manual transmission car without touching the brakes and there’s such a thing as going uphill. I think the bolt solution of an accelerometer gives the best indication that a car is slowing down quickly.

It's a US problem probably, everywhere manual transmission was or is a thing everybody expects cars to slow down. Also: keep enough distance, pay attention and this is never an issue.

> keep enough distance, pay attention and this is never an issue

That isn’t how safety regulations work. Driving is a cooperative affair and giving more information about your intend to the drivers around you is a good thing.

I mean if all it took was “keep distance and pay attention” why bother with turn signals?

And as a side note: “self driving” cars still haven’t completely figured out the right algorithm for turn signals. I watch enough “AI Driver” on YouTube to see him call out how Tesla’s turn signaling is sometimes pretty erratic. It is probably a pretty hard problem when you think about it.

Showing brake lights is much easier. Got a deceleration over a certain value? Well, show them lights!


> I mean if all it took was “keep distance and pay attention” why bother with turn signals?

Have you met BMW drivers? /s

You should be practicing defensive driving whenever you drive. It's what's taught nowadays and has been for a long while.

You should expect others to be driving at their worst. No turn signals, broken tail lights, erratic lane change, abrupt braking or acceleration, the whole shebang. Expecting the worst is the best way to ensure your own safety on the road.


I laughed at the mention of BMW drivers because it's so spot on. I agree that defensive driving is always the way to go, but at a certain point you need information. TC essentially goes from 100km/h to stopped in less than 15 seconds without any obvious visual indication. That's coming to the point where there's almost no defensive driving possible and a deadly highway collision becomes more likely than not.

It's the same in any situation. Always assume people will do completely insane things and prepare accordingly, but that doesn't mean we should make things more dangerous for no good reason. For all the truth that more chaotic roads make for more attentive drivers, they also still make for more accidents.


There are two different problems.

One is educating drivers to be better. Drive more predictably, drive more defensively, drive safer in adverse conditions, etc.

The other is designing the driving experience as a system to deal with humans. Systems that don't anticipate human faults are brittle and, frankly, shitty. Just like you can't expect 100% of users to be phish resistant, you can _not_ expect perfection from drivers, and you _must_ build in compensations to deal with it. That means well designed roads, good brakes lights, and even brake lights that account for one-pedal driving.


Adding new indicators is an obvious and inoffensive thing to do. I don't think anyone is claiming we shouldn't build cars to be safe.

It is, however, horrifying that there are drivers so incompetent they don't even know what engine braking is. It is hardly an advanced topic. Any driver who doesn't know cars can decelerate without the brake lights coming on has no business being on the road.


Yes. But the point is more information is good. You can be the most observant most defensively best trained driver, and more predictable information is still good.

I don't understand the motivation of arguing an objectively provably demonstrably good thing, because some other thing is also good :-).


Sure, more information is good, but you shouldn't expect that information. People don't use DRLs and turn signals all the time. Lights break. Electrical connections degrade. Old cars almost never keep up with advancements in signaling technology.

All the signals in the world wouldn't help people deal with someone that refuses to use their indicators or doesn't maintain their old car, and that is the expectation you should have when driving: the worst possible drivers around you all the time.

You can deal with practically any situation as long as you're keeping your distance and staying observant. Regardless what the people around you are or are not doing.


Agreed. Fully.

And yet precisely none of that, to me at least, is persuasive reason to NOT put the most effective possible light and information on cars coming out of the factory :-)

(And don't even get me started on red vs amber rear turning signals ;)


In my neck of the woods at least, driving safely is impossible. Nearly every major road is over capacity and getting worse. Most drivers are half-trained at best. Constant traffic jams fray the nerves. A lack of public transportation means people are on the road for longer than they can focus. The best selling vehicle is the Ford F-150. Leaving adequate following distance just results in someone taking that as an invitation to swerve a quarter second in front of you.

We should develop every technical solution we can, but the reason American drivers are in so much danger is social.


Well yes, but engine breaking is a lot weaker then regen braking can be, so there really needs to be break lights id deceleration is more than one would reasonably expect engine breaking to achieve

Really depends on the car. There are cars with very little inertia, if you let go of the gas pedal quickly they will get quite jerky. In those, you really have to depress the clutch before completely letting go of the accelerator.

What kind of a car is that? I have 1.6l hyndai ix20 and I can let off gas at any time with zero issues....

> keep enough distance, pay attention and this is never an issue

If this was helpful, far fewer crashes would happen on roads today. People are imperfect. They're distracted, they have slow reaction times, they have poor vision, and they drive in poor weather.

As the video states, regulations should definitely be altered to require car manufacturers to show brake lights no matter how deceleration occurs. It's just one additional way to prevent collisions.


Keep enoug distance to be safe and someone will just change a lane to be in front of you, removing any safe distance you decided to keep

Then slow down a bit and make more space. Traffic flows better when people have room to change lanes.

That's my complaint, if I do someone just fills that space because their lane is going 2cm/minute slower than mine

Then slow down a bit and make more space. There's no trophy at the end.

Which causes next car to get in front. Still no space, nor more safety, and also triggering morons to do extra maneuvers and potentially fuck up more.

Do you lack ability to think about the results of the actions ?


Going uphill isnt a problem because the car behind you is also going uphill. The point of brake lights is to warn driver's behind you they need to brake, which is why it needs to be a measure of negative force being applied by the car, not just the resultant acceleration vector.

I've thought about putting a manual brake light switch in my manual transmission car because of this, but I'm also that guy that engine brakes more than actual brakes. 6 years of owning this car and I've never worn out a brake pad.

Wouldn't it be easier to just put some light pressure on your break pedal while you're engine breaking? My older Acura definitely lights up the break lights well before actually engaging the brakes.

Much easier, but not as technically interesting :) And I do pedal it if I think the driver behind me is risky, though I'm not sure where the light engagement is relative to brake engagement.

That foot is on the gas pedal still in a lot of cases

It's so nice to not have to ride your brakes going downhill, especially when you also have to turn at the same time. (I miss my old manual car)

Newer manuals are really nice.. mine has hill hold, which some people hate but I enjoy. And supposedly engine braking uses no fuel.

Unless you have an older carbureted vehicle, engine braking should use no fuel:

https://en.wikipedia.org/wiki/Engine_braking#Limitations

Just do a decent job matching revs on the downshifts so you don't trade brake dust for clutch dust :)


Not all hill holds are created equal. On my car, the brakes release as soon as you start moving forward after hill hold, or after 1.8 seconds. On my friend's subaru, the brakes hold for 2 seconds period. That means if you can do a hill start in less than 2 seconds, you drag the brakes every time.

15 civic SI. Hill hold is 1.2 seconds I think and no early disengagement. But it's rear brakes only. You can't disable it that I know of.

Why would you need an accelerometer when you've got perfectly good tacho pulses from the ABS rings?

AFAIK ABS modules already have accelerometers. If I had to guess it falls into the domain of the ABS module, and it's just easier to talk about it as an accelerometer, when it's really using multiple inputs (wheel speed sensors can fail).

I hope whatever new regulation is intelligent enough to not consider the deceleration from coasting up a hill to be part of triggering the brake lights. Brake lights should only come on when there is deceleration beyond what you'd expect from a naturally rolling car.


One would assume it's not a difficult process to model.

If the driver presses the brake, put the brake lights on.

If the driver lifts off the throttle in heavy regen mode and the vehicle slows faster than <some amount> put the brake light on.

If the vehicle slows faster than <some amount> with the throttle "off idle", do not put the brake light on.


In theory you can, but 99% of the time you'd still generally use the brakes for any serious deceleration. On the other hand with one pedal driving, regen braking is used 99% of the time.

Solved problem. My Teslas have always activated brake lights based on deceleration. My wife's Bolt does as well, though it does have a glitch -- once it comes to a complete stop, it turns off the brake lights if you don't have your foot on the brake. Tesla gets it right, keeps the brake lights on until you start to move again.

That's covered in the video - GM uses an accelerometer to detect braking. Once you stop decelerating, it turns off.

GMs now incorrectly turn the backup lights on when in park.

Bastards.


Noticed this the other day when behind an Ioniq, I knew immediately it was because of regen braking. I love Technology Connections for taking it a layer deeper and going into the regulations and industry experts.

Sounds like an Ioniq specific oversight and bad design not a general problem with break lights

If you watch the vid, he starts out with the Ioniq's problem but then goes into discuss how their implementation actually meets the regs and therefore the regs are bad.

Regenerative braking should trigger brake lights if it's strong enough. The car knows your brake pressure and it knows how much regen torque it's applying. Just sum those two and turn on the light at the same threshold as before. OTOH there may be a federal regulation that doesn't permit that. I haven't studied FMVSS in a while.

Did you watch the video? Because that's exactly what he says. The whole issue boils down to the fact that the regulations are outdated, not that there are conflicting regulations.

Worth noting is that even if the regulations could be worded better, nothing is stopping Hyundai from doing it correctly. It's not forbidden by regulation, most other manufacturers get it right.

Regulations are there to help in cases like this. If the regulation is failing to accommodate modern vehicles, they need to be changed. Trusting companies to "get it right" on their own and act in the best interest of peoples' safety leads to unnecessary deaths. Seriously, back when seatbelts were invented car manufacturers were almost completely against it. They lobbied against regulations to implement them.

Luckily, at the very end of the video, it's actually revealed that a couple of months ago the new regulations were put into action (about a decade late)


EU regulations were updated. Not everywhere.

While they have these cars in the shop for the theft related recall* they should update the computer to fix the brake lights too.

*I have no idea if this specific car is recalled.


> *I have no idea if this specific car is recalled.

No, the only affected cars were ones that used a traditional key. Not the push-button cars, which necessarily excludes all their EVs as well.


From Tesla (note this is for the Model Y, but is true for all models):

--------------- If regenerative braking is aggressively slowing Model Y (such as when your foot is completely off the accelerator pedal at highway speeds), the brake lights turn on to alert others that you are slowing down. ---------------

I had kind of assumed all electric cars did that, but I guess not.


This was seen as a problem in Formula 1 when they moved to hybrid engines - and so when a Formula 1 car is engaging regen during braking for a corner the red light they have on the back of the car flashes to let any car following know that they are likely to stop faster than if just using brakes alone.

Wow I actually drive a Bolt pretty often in a shared EV fleet and I always wondered what the hell the "L" on the shifter was supposed to mean.

The biggest issue with the brakes on the Bolt is the way the brake pedal software works. In every other car I have ever driven, a steady pressure on the brake pedal will bring the car to a halt. But on the Bolt a steady pressure on the brake produces (inexplicably) proportionally less braking as speed falls, meaning to bring the car to a normal stop you need progressively more force on the pedal. It makes the car feel as if its brake lines are full of air, even though it's all computer-controlled regenerative braking and the hydraulic system only engages at the very last moment.


Regenerative braking is disabled at very low speeds as it becomes inefficient/ineffective. That's when the mechanical brakes need to be applied. Some cars might need a bit more pressure on the brake pedal for such. In a Prius (2nd gen.) this transition is barely noticeable (and I assume was the reason for some of those "sudden acceleration" claims years ago).

This all applies of course only to the soft-braking when coming to a stand-still. When braking hard, the mechanical brakes are applied at high speed as well.


While brake lights are a fine and necessary signal, I find myself wishing that cars would display speed & acceleration deltas compared to the vehicle in front. Brake lights do not convey how much deceleration is occurring, so it's not immediately obvious how much time there is to impact.

As an example, I see lots of drivers do minor braking on US highways, where their small braking corrections will cause their tail lights to flash for a brief moment. These minor corrections become noise that drowns out the signal of significant, safety-related braking. If my vehicle could warn me that their rate of deceleration puts our vehicles on a collision course... now that would be very useful.


Some do: In ACC mode my 2013 Volvo uses radar to "lock" on the car in front of me. If it is slower than the ACC set point, its speed is indicated in the instrument panel.

It looks like this: https://www.motoreport.de/wp-content/uploads/2014/01/IMG_296... Set point is 110 km/h (green marker), the car in front is going 80 km/h (solid white line indicates the speed difference). The small red triangle indicates the speed limit of 80 km/h. Obviously the ACC is also cruising at 80 km/h, so most indicators are focused there.

And you're right, even with the limited form (single car in front of me on the same lane, data only shown if slower) this is amazing.


I think Tesla solved this problem.

https://www.tesla.com/ownersmanual/modely/en_eu/GUID-3DFFB07... "Note If regenerative braking is aggressively slowing Model Y (such as when your foot is completely off the accelerator pedal at highway speeds), the brake lights turn on to alert others that you are slowing down."


No, Hyundai needs to rethink how brake lights work.

Putting yet another mark on the board for things Tesla engineers figured out nearly a decade ago that other car companies haven’t.


Nissan figured it out in the original '11 Leaf before the first Model S was even delivered. Other companies have figured it out as well, it would seem Hyundai just dropped the ball on their design.

> before the first Model S was even delivered

The original Roadster came out in 2008.


The point was regulations should be updated. And comments said other manufacturers got it wrong.

My VW ID.4 does this correctly: the brake lights come on whenever the car slows down, which could be because I pressed the brake pedal or because I released the accelerator pedal in "B" mode.

The software on this car sucks really badly in too many ways to describe here, but this part they got right.


Thanks for posting! I was just thinking about how I was going to confirm this on mine.

And, yeah, the software is hot garbage.


I would argue that, no, it's Hyundai that needs to rethink brake lights (though I suppose we might write some regulations to ensure consistency). I have an Ioniq 5, and this is one of the two reasons I don't use one-pedal driving (the other is that the car always uses both motors in one-pedal mode, therefore reducing range). There are many discussions on https://www.ioniqforum.com about the subject. There are also three levels of regeneration, and apparently the highest level (Level 3) also does not light the brake lights appropriately, whereas Level 2/1 and "auto regeneration" apparently do.

Comments said other manufacturers got it wrong also.

I am in the middle (south of SF tonight) of a road trip from Palm Springs to Seattle in an electric campervan. This is my first EV experience and the regenerative braking is very strong. I immediately considered the brake light issue and I’m not sure what to even do. Applying the service brake seems wasteful, I have limited range and want as much regen as possible. This should definitely be addressed. Brake light activation should be based on deceleration, not pedal engagement.

For motorcycles (with combustion engines) where engine-braking can be quite effective due to the high displacement/weight ratio, there exist (but afaik isn't widely known or deployed) custom brake-light switches which can additionally be activated by chopping the throttle or downshifting.

I never had such device and relied a lot on engine braking (and on followers to pay attention to the distance, rather than my brake lights). Every once in a while I surprised a tail-gater, but luckily none hit me yet.

I rather think we need to rethink driver education.


I have been driving my hybrid under the assumption that my brake lights triggered when ever the car used regenerative braking. Guess I will have to be more careful.

Have someone drive behind you and ask them if your brake lights come on during regen braking, or do what TC did and attach a camera to the back of your car.

Based on other comments in this thread, your hybrid probably triggers the brake lights during regen braking. It seems only Hyundai got it wrong.


Driving a Volvo BEV, they got it right. One pedal driving + slight regen doesn't light it up but anything harder where you can feel the deacceleration in your seat, the brake lights illuminate.

I haven't driven a lot of other EVs but Volvo put a lot of thought into their one-pedal mode and it's fantastic. It's incredibly precise.


If you're not driving a Hyundai, it's probably working as you expect.

I'm driving a Honda. I looked it up and people online say that it has the same issue. However, I plan to observe for myself.

Now I'm curious about my Volvo. I do most of my freeway driving with the adaptive cruise control adjusting my speed to match the flow of traffic. I have no idea if my brake lights are triggering when the ACC slows my vehicle due to a slower car in front of me. I certainly am not touching the brake, unless I need to stop more quickly.

Given Volvo's reputation for safety, I'm assuming they got it right but that is purely assumption on my part.


My guess is that most decent TACC implementations won't activate the brake lights when slowing for a car in front, unless that car cuts you off. Generally the deceleration should be slow enough that it falls under 'engine braking' levels and therefore unworthy of a brake light.

Self-driving cars should flash their brake lights when they are entering an intersection with the intent of getting a view around obstructions before advancing further. That, according to California DMV reports, is when they get rear-ended. (They don't seem to get hit by cross traffic, so they're doing it right.)

People have been decelerating without brakes or brake lights forever with manual transmissions.

It would have definitely been unreasonable to require some kind of accellerometer system just for that in the 50s, but it's probably not asking too much today and probably more justifiably neccessary with one-pedal mode.


Does anyone know how this works in Ford EVs/hybrids? I have a PHEV CMAX and have wondered about this. I've never had an issue with people honking or getting close to rear-ending me, so I assume the lights do come on when it's decelerating rapidly.

Fusion Hybrid '18 here -- no brake lights on regen activity, just the dash icon that you can enable to see that it's regenning. It's not harsh regen at least on the full hybrid (even in Eco w/ Grade Assist enabled) until you slap it in low for engine braking.

I drive in L almost all the time.

I previously posted this at https://news.ycombinator.com/item?id=36071933 I'm not sure why it didn't deduplicate.

That submission is marked as [dead]

Corvair, Pinto, Suzuki Samurai, seatbelts, head restraints, offset crash testing: Safety regulations are written in blood. Only when enough people have died from manufacturers skirting technicalities will anything change.

Are there statistics anywhere on which cars are most likely to be run into the back of? I'm sure insurance companies would be able to collate the data. It would be interesting to see how much of an effect this has.

Seems like a software problem that they will hopefully fix. Teslas do this already.

Related tangent: Lemon law lawyer on Youtube: https://www.youtube.com/c/stevelehto

I'd like to show people behind my accelerator / brake / clutch state. Would be cool if they had the information so they could react appropriately.

Clutch state would be lost on at least 95% of drivers in the USA.

Honestly, I'd wager that half the people driving a manual car that's not in a sports car doesn't even know what the clutch is actually doing. They just know they need to press it while shifting and hold it halfway out for a second or two when coming from a stop with no concept on why they have to do that, other than "That's what you do in a manual transmission".


I have seen some cars that have pulsing brake lights to indicate aggressive braking. Don’t know if this was aftermarket or not.

Or just huge bargraph at back showing acceleration/deceleration of the car.

Yeah, I was thinking of that. Sounds like it would be neat. I wonder if it is legal.

fun fact: break lights are red because red light disperses / scatters less than other parts of the spectrum like yellow or green f.x.

and the sky is blue because blue light disperses the most - so everything blue in the sun's light gets dispersed into the sky while the color of the sun is what's left of its full spectrum minus blue.


This is part of a larger class of problems, which is "car manufacturers are not taking some quite simple steps to make the roads safer".

I think this could be solved with a small fee, payable by a vehicle manufacturer, whenever one of their vehicles is involved in a road traffic incident.

Perhaps $100 would be enough for them to keep an eye on their fleet of cars on the road, and do basic things to minimize accidents, like for example tyre wear alerts.


Tesla is now in the insurance business. So it has a direct incentive to prevent crashes, as well as to make the car cheaper to repair.

They also have the reverse incentive if their cars get into crashes which are other people's fault.

They get to profit from very expensive repairs, paid by the other guys insurance.

Making sure the brake lights work more effectively than the regulations require, reducing rear-endings of tesla customers, would then reduce profits...


Billing a company for something that is entirely outside their control (post sale) isn't going to fly.

But you could maybe publish safety figures per manufacturer on the road, and make the scores for a new car incorporates the manufacturer score too - therefore effectively giving a competitive advantage to manufacturers that make cars which turn out to be safe.


That has added advantage of preferring cars that are actually safer and not cars that filled some safety checkbox.

How does one make a video about "electric cars" without considering that the most popular electric car brand by far (Tesla) has solved this issue years ago?

There are more Teslas on the roads in the US than all other electric cars combined.


Author mentioned that other car manufactures solved this issue. No need to mention Tesla specifically.

What do other manufacturers do?



Applications are open for YC Summer 2023

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: