How much do electric car batteries degrade? | Hacker News Reader

How much do electric car batteries degrade?

sustainabilitybynumbers.com

135 pointsxnx276 comments

> At that point, a Tesla will have more than 80% of its initial capacity, and in some cases, even more. So people will probably give up their car, well, well before the battery gets close to becoming a burden.

I looked into the secondhand EV market (in Norway). In doing so I read quite a bit of academic research to figure out the lifetime of an EV. Apparently the 80% capacity is the accepted end of life for an EV battery:

"For batteries, 80% of the initial capacity is referred to as the point after which it tends to exhibit an exponential decay of capacity and is considered an unreliable power source after this point for EV application" [1]

So, the Tesla the article talks about won't be much good, or at least not for very long.

[1]: https://doi.org/10.1109/ACCESS.2023.3271287

That's statistical. A small number will start their exponential degrade at 80%, but most won't. Some might get to 60% before they start it. So if you're at 83% at 200,000 miles you don't really know whether you have 50,000 more or you have 200,000 more on the battery. And "exponential degrade" doesn't mean it's particularly fast. It means it's faster to degrade from 80 to 60 than it is from 100 to 80. You're not going to get 500,000 miles out of driving until the battery hits 60% but you might get a substantial fraction.

The rest of the car almost certainly doesn't have 200,000 more miles in it, so who cares?

Why do you think a vehicle with 200,000 miles on it doesn't have 200,000 more miles in most of the components? It isn't remotely difficult to imagine that being possible. One of my ICE vehicles was 299,648 miles when a component in the transmission gave out. The only reason I didn't repair it was around a decade earlier someone had bent the frame. If I had been willing to replace that part, I would still be driving it today.

People replace vehicles because they want and can afford replacement ones, not because they are mileage limited.

> People replace vehicles because they want and can afford replacement ones, not because they are mileage limited.

That isn't quite true. It is often because it is uneconomical to keep.

The issue with many older vehicles is that it just starts costing more to repair it than the car is worth. You are constantly putting in good money, after bad and the the costs keep going up over time. At that point it is often better to buy another vehicle.

I could have kept my 2007 Vauxhall Astra going for another 100,000 miles. The issue is that the car was worth £600-1000. Each repair was costing upwards of £300.

I could buy another second hand Vauxhall Astra with the amount of money I was spending on the car yearly. It just wasn't worth it.

This is just a bad application of sunk cost. The reason why you spend 300 pounds to repair a 1000 pound car is because it is a known item. If you buy a used one for 600 pounds, it may need even more repair than the one you already have.

Repairing a known item makes sense when the costs are small compared to a replacement. I have old bicycles that are worth nothing and I am quite happy to spend maybe £50-£150 in components to fix. Costs are much lower than buying a new good bike or trying to find another good second hand bike (which I would need to fix anyway).

However that frequently isn't the case with car. It is £300-1000 every time I need to get it repaired. This happens at an increasing frequency as the car ages. This can then add up to several 1000s quite easily.

e.g In the last year I owned that Astra I had to to fix the following issue (all costs include labour):

- New Clutch (and it needed to be towed to the garage as the clutch wouldn't engage). £1000 for clutch replacement, £150 for the tow. I also needed to rent a car to drive to a funeral. That cost me another £100-200 IIRC. So we we are up to £1350.

- Car would randomly go into Limp mode. New sensor cost £300. Vehicle was never really fixed. It just went into Limp mode much less, so there was another problem somewhere. Which would need another trip to the garage.

- Service. This flagged several issues with the car. All these small repairs was another £800.

I am already upto £2,450 and I know I've forgotten stuff. I bought the car for £4000 originally.

If it was something like a classic barn on wheels Volvo which are bullet proof, or a classic Rover like a Rover P5. I might be willing to keep dumping money into it. But it isn't, it is a Vauxhall Astra.

With 4k I would to go for a Focus or Fiesta, I think you're always risking trouble with a Vauxhall? The Focus of around 12-14 years ago, particularly, was seriously over engineered which is why you still see so many of them around.

I had one for 4 years, approaching 200k miles, it came to grief at the hands of a 95 year old lady driver. Her car was wrecked, the Focus stood up well but, alas, even the minor repairs required were deemed uneconomical.

No help to you, admittedly, but if it assists anyone else avoid a similar issue...

The Astra G and H 1.7 cdti variants were really good. Newer ones not so good from what I have heard.

Ford Focus around the same time period are indeed decent. Going to try to get one at some point.

If the rest of it seems solid, then it can be worthwhile. But if you start spending 300 every 3 months then you're just throwing money down the drain.

Again that's statistics. Some 200,000 mile vehicles have 200,000 more miles on them. Some don't. If you can live with that uncertainty, you can save a lot of money.

Most Diesel Engines (even ones well looked after) won't last after 300,000 miles and will either need to be rebuilt or replaced. A replacement Diesel Engine is several thousand pounds and that is before labour.

Then other things like hoses, anything rubber will perish and will be replaced. Everything starts going at some point and the cost mount up quick!

Where do you keep getting all this information from? Yes, there are some engines that suffer catastrophic failure well before 300,000 miles. But it isn't even most. I have rubber parts on vehicles that are decades old with no sign of degradation. Vehicles are not made of compost.

On a regular car 300,000 miles is like 20-30 years of use (approximately 30 miles a day). How many cars you see on the road today that are over 30 years old? I barely see any vehicles over 20 years old.

Things wear out on an engine over time, you lose compression over the time and thus power, you can have other components that will reach EOL. A lot of this starts happening on vehicles well before 200,000 miles. I know it does, I've had to deal with it. I am having to replace a knackered turbo on one of my vehicles, and I had to replace the belt tensioner and replace the timing belt, I've had to reseal the sump.

Rubber perishes over time. Regular maintenance and what conditions something subjected to can improve the life sure. I've replaced plenty rubber bits and pieces because they've perished. There is a reason they offer rebuild kits for some components, because those rubber / plastic parts go bad after a while.

These are just facts.

> On a regular car 300,000 miles is like 20-30 years of use (approximately 30 miles a day). How many cars you see on the road today that are over 30 years old? I barely see any vehicles over 20 years old.

I drive a '91 GMC Sierra. It has 138k miles on it right now. It was made in 1990, making it 35 years old.

> Things wear out on an engine over time, you lose compression over the time and thus power

I've not put it on a dyno, but qualitatively I'd say the 5.7L (350ci) v8 in my truck has 90% or more of its original power. The miles are much less important than how it was used. Mine belonged to my grandfather, so it wasn't exactly street raced.

> you can have other components that will reach EOL.

Sure. This is called "maintenance". It's an ongoing cost - and a compounding one, at that, if you don't stay on top of it. If you do, an older vehicle can be kept in excellent working order for not much money at all.

> A lot of this starts happening on vehicles well before 200,000 miles.

Sure. Most modern vehicles require their first major service at 50k-100k miles.

That's not a defect. That's how they are designed. Some parts wear faster than others intentionally, because they are meant to be replaced.

> I know it does, I've had to deal with it. I am having to replace a knackered turbo on one of my vehicles, and I had to replace the belt tensioner and replace the timing belt, I've had to reseal the sump.

The turbo failing is the only thing here that raises an eyebrow for me; everything else is just standard maintenance.

Modern turbos are doing a lot of work, and depending on the vehicle are often included to meet environmental regulations.

---

More to the point, ICE vehicles are well-understood and can be both maintained and repaired by novices without significant risk. I recently bought my wife a 2019 F-150 Platinum. It was in excellent condition overall - interior and exterior are basically perfect - but had a massive oil leak around the timing cover. It had 97k miles on it.

We paid $22k for it. I put $1,500 in parts in it: replacing the $20 gasket meant tearing the front of the engine down, so I went ahead and replaced the cam phasers and all timing components at the same time. I expect it to last another ~60k miles before the next major repair.

I've already budgeted to replace the twin turbos and transmission over the next few years. Even with those costs amortized into my budget, and including the initial cost of the vehicle, I'm paying about 1/3 of what it would have cost me to buy a new vehicle of the same quality.

What's more, it gets ~24 MPG city and ~28 MPG highway. The only modification I've made to that effect is that I added a (~$300) soft tonneau cover to the bed.

My '91 GMC gets 12.5 MPG city... but its retail value is also ~$10k at most. I'd have to drive a lot of miles on a regular basis before it would make economic sense for me to replace it with something more efficient - and even then, I'd still want an older truck, because I don't particularly want to fill our shiny F-150's aluminum bed with gravel but have no problem doing the same with my GMC's steel bed that already shows 35 years of use.

From an environmental perspective, I bet I could drive that truck for the rest of my natural life before I consumed more resources or exhausted more pollutants than would be required to manufacture a new one.

> From an environmental perspective, I bet I could drive that truck for the rest of my natural life before I consumed more resources or exhausted more pollutants than would be required to manufacture a new one.

Unless you drive much less than the avarage this is not true, engineering explained [1] did a video about this, and comparing a 24 MPG to a 35 MPG car you produce more CO2 with the 24 MPG car compared to a new 35 MPG car (including the manufacturing) considering your GMC has much worse fuel efficiency that break even point would be much earlier. This is not even considering the impact of recycling... So the best thing for the environment you could do is buy a new more fuel efficient car.

[1] https://www.youtube.com/watch?v=L2IKCdnzl5k

Sorry seems I somehow messed up my post, the overall CO2 budget of the new more fuel efficient car is better after 7 years.

I do in fact drive much less than average; I work from home and mostly drive when doing "farm work".

I am just going to leave this reply I left here on a sibling thread as I addressed most of this there.

https://news.ycombinator.com/item?id=44951308

It is often not worth continuing to maintain a vehicle after a certain point.

I can do quite a lot of the jobs. But I don't have really anywhere to work. I live in the UK and I cannot even rent a garage within 30 miles. So I have to work in my car parking space.

Working on your car yourself, isn't hard if you have the time and space. I don't have space really.

> The turbo failing is the only thing here that raises an eyebrow for me; everything else is just standard maintenance.

I don't think the previous owner maintained it properly. It lived a hard life on a farm.

sidewndr46 says " One of my ICE vehicles was 299,648 miles when a component in the transmission gave out."

I'm guessing it was a Lexus?

I drive my vehicles for 300-500k miles. The drive train should be the only thing that wears out and that can be replaced with used drive trains from a wrecking yard.

What are your breakdown patterns ands how do you go about defending / being prepared? I bet you have a lot of useful advice.

Not the GP, but I can answer. My current vehicle is 35 years old. Prior to that, I had a '00 Jeep Wrangler for 15 years. Before that was a selection of older vehicles: an '88 Toyota pickup (that I dearly miss...), a '97 Ford Ranger, a '99 Dodge Dakota, and a '98 Honda Accord.

I keep a small toolbox in my vehicle. It's mostly inexpensive hand tools, but I include a Milwaukee M18 Fuel impact and a set of sockets for it - super handy for changing tires.

In my 25 years of driving, I've broken down probably a dozen times total. Of those, only twice have I required a tow - and one of those was in my wife's Kia Sorento, which we bought new.

Easily 90% of the usage of the tools I carry ends up being on other people's vehicles. I can swap a wheel on the side of the road in <10m with a bottle jack and a battery-powered impact, with no real manual labor involved.

Other breakdowns I've had in the past were things like the serpentine belt breaking or a coolant hose coming loose. Those are five minute fixes if you have access to a parts store. When the belt snapped on my Jeep in the middle of the night on a trip a few years ago, I used an pair of my wife's leggings that she had packed to get us home. I just tied them by hand, bypassed all the non-essential stuff on the motor, and drove the ~50 miles back home to deal with it the next day.

The only time I have ever purchased a brand new vehicle was in 1987 (Mustang GT), 2000 (4Runner). Still have both and drive the 4Runner regularly. Every other vehicle (2007 Avalon, 2010 Venza, 2006 Lexus IS350) was purchased used and all are daily drivers. Since the 4Runner, my family only drives Toyota and that is a big part of why I can keep a vehicle for 350-500k. Most of the driving is highway miles, so that also helps.

I service my own vehicles and do a lot of preventative maintenance and inspect the condition of my vehicles regularly, it's become a hobby since I enjoy working on them. I adhere to a strict schedule of maintenance and inspection. Just flushing and filling fluids (Brake System, Transmission, Power steering, Rear Differential, Coolant and Oil) can greatly extend the life of the power train. I have never had a break down where I needed to call a tow truck, Each vehicle has a small jump pack in the trunk, alternators, batteries can be tested for signs of degradation, why wait until they fail, replacing worn parts before they cause further damage. I always try to rebuild the OEM part since it is of higher quality then the replacements, like an alternator or starter. If I do need to get a replacement, Rockauto carries many OEM parts. I just replaced the original water pump on the Avalon, it had 160k miles on it. The 4Runner has approx 350k miles on it and many of the parts (alternator, starter, etc are original.

He’d answer but he’s busy driving

Every vehicle comes with a service schedule. If you know nothing about maintaining or repairing a vehicle, follow what it suggests (it's in your glove box). It might cost more in the short term but save you more in the long term. A friend does that and his vehicles last forever and reach crazy millage (he drives a lot as a regional sales manager).

> The rest of the car almost certainly doesn't have 200,000 more miles in it, so who cares?

This what I find curious. It used to be cars were largely limited by their main driveline components. Now those components have been simplified. In a lot of the BYD cars it is literally the exact same driveline so for them it is almost commoditised.

My part of the world sees no snow or salt. So if the body is fine and the driveline is fine....

Why can't I run a car for a million miles and simply replace coils, struts, hubs, bearings, carpets, seats, the-tiny-electric-motors-that-drive-windows-and-seats-and-mirrors etc?

IE where is the electric version of the old landcruiser series?

That's just "any electric car" as long as you're willing to replace the battery. Since the only thing in the drive train that moves is the electric motors (and contactors), any EV is way more mechanically simple than any internal combustion car.

The non reliable stuff is all the OTHER wiring/screen/software.

I recently drove in an early Tesla with over 400K kilometers on the clock. Still runs quite fine and the range seems longer than my Zoe's.

> That's statistical. A small number will start their exponential degrade at 80%, but most won't. Some might get to 60% before they start it.

The paper that underpins the 80% claim is paywalled and I'm cheap so don't have the numbers to hand. But in the abstract sense being statistical could go the other way too - i.e. some might get to 90% before they start their exponential decay.

> And "exponential degrade" doesn't mean it's particularly fast.

True, exponential doesn't mean much other than e^n where it's implied that n > 1. However the value of n does matter, for example a decay rate for n=100 is quite different than n=1.1. Again, we're talking in general terms as I don't have the value of n for this. For better or worse, I'm taking the paper at face value when it talks of end of life at 80% - however it has been peer reviewed and is in a "proper" journal so to my mind that adds some credibility.

It's worth mentioning that the current DoE guidance on lifetimes is: "12 to 15 years in moderate climates (8 to 12 years in extreme climates)". Of course that guidance is also based on a statistical analysis that doesn't seem to account for modern battery management systems [1]. There are additional factors at play that govern battery degradation, including things like: battery chemistry, environment, EV charging habits, etc.

Ultimately the real test of EV battery life will come with time. It'll manifest as the change in average age of scrapped vehicles with large fleets of EV's on the road. In Norway right now it's about 18 years (for a mostly internal combustion fleet) [2]. However as of this year all private vehicles sold must be zero-emission - so in 10-20 years we'll have some interesting data [3].

[1]: https://web.archive.org/web/20250530000446/https://afdc.ener...

[2]: https://www.ssb.no/en/statbank/table/05522/tableViewLayout1/...

[3]: https://web.archive.org/web/20250530115356/https://www.regje...

Edit: It'll manifest as the change in average age of scrapped vehicles [in countries] with large fleets of EV's on the road.

> For batteries, 80% of the initial capacity is referred to as the point

The publication cites sources from 15 years ago for this "fact" [0]. That's ancient history in the context of EVs (even before the first reliable mass production EV - Tesla's Model S - was initially released). As a practical matter, the article points out that most EV manufacturers (Tesla included) warranty their batteries for at least 70% capacity at timespans near a decade, which would bankrupt them all if EV batteries just up and died at 80%.

0) [35] O. Erdinc, B. Vural, and M. Uzunoglu, ‘‘A dynamic lithium-ion battery model considering the effects of temperature and capacity fading,’’ in Proc. Int. Conf. Clean Electr. Power, Jun. 2009, pp. 383–386.

[36] K. Smith, T. Markel, G.-H. Kim, and A. Pesaran, ‘‘Design of electric drive vehicle batteries for long life and low cost,’’ in Proc. Accelerated Stress Test. Rel. (ASTR), IEE Workshop, 2010, pp. 6–8.

It is an older paper, though that in itself isn't a reason to discount this (or any science). For example, the DoE still uses an 2014 predictive model to estimate longevity of today's EV batteries at "12 to 15 years in moderate climates (8 to 12 years in extreme climates)" [1]. However as the DoE says, battery longevity depends on a bunch of factors e.g. chemistry, charging patterns, etc.

To the practical matter - yes, EV manufacturers are very careful with warranty periods. Anecdotally, an acquaintance had a Tesla for 8 years. 6 months after the 8-year battery warranty expired the battery ceased working. The details were a little unclear (it was explained in broken English/Norwegian). That said, anecdata carries little weight. What we need is more peer-reviewed research to update our understanding of battery longevity. Until we have that we need to rely on the existing published knowledge... otherwise anyone can assert anything and we learn nothing.

[1]: https://web.archive.org/web/20250530000446/https://afdc.ener...

I kind of wonder if the tipping point of "exponential decay" might be where a battery starts being charged and discharged more often to reach daily range and maybe outside the 20-80% window you need for a healthy battery.

In other words, if a battery is new and has 200 miles range, but is driven 100 miles a day, it will stay between 20-80% charge each day when charging.

but at 80% capacity, 160 miles range, it must be either charged above 80% or discharged below 20% each day which is unhealthy for the battery. (either 80%->17% or 83%->20%)

as soon as it starts getting out of healthy range I can see how it can degrade faster.

But cars that stay in that range will have a much longer lifespan.

As another comment mentioned, the cited studies are more about how the battery chemistry changes over time, not how people use the EV. That said, EV manufacturers definitely care about driver habits too, since it affects warranty claims and how they configure their battery management systems.

I think modern EVs are set up to charge to a certain percentage of capacity rather than a specific range, kind of like how smartphones do it.

Your point about depth of discharge ( 80%->17% ) makes sense though, since the battery system can't really control someone's driving routine.

> I’d say that it’s more likely to be the perception of battery degradation that pushes the value down, not the actual degradation in reality.

Why guess? This is data that is almost certainly aggregated back to the manufacturer and could be available as a published report. The fact you don't see this report I think is indicative of the reality.

> Pessimism about battery longevity is giving us all cheaper second-hand EVs

The seller sets the price. Not the buyer. You should ask why the seller is willing to let the vehicle go for a lower than expected price for a given number of miles.

> had lost just 15% of their capacity, on average

There are no average batteries. The used vehicle market doesn't work on averages, it generally works on worst cases, particularly in an as-is (no warranty) sale.

> in other words, there was no active cooling of the battery

As it requires a four way valve. That's a common failure point in EVs.

> Many manufacturers provide long warranties for their batteries

Do those often persist through private party sales?

I'm not trying to be mean, I think EVs are great, but hybrids are still obviously better, and the market is far more complex than this author would like to acknowledge. I dislike articles that start with a conclusion and then spend pages trying to justify it. The data to actually answer this question was available but completely unused here. I did not find this convincing or informative.

Think of it this way. There is a segment of the buying market that is hostile to these vehicles by default. Waving your hands with no data to back it up only makes it worse.

Anyways..

I like ICE cars. Someone literally had to pry the keys to my Integra out of my cold, (half-)dead hands. I like EVs: My Bolt's great. I will never own a hybrid, after working with my parents'. They are simply the worst of both worlds: With an ICE, if it breaks, I can probably fix it myself, likely cheaply. With an EV, there are vanishingly few moving parts to break. With a hybrid? All the failure prone parts of an ICE, packed into a smaller engine compartment because they had to stick an entire scaled-down EV in there on top of the ICE parts, making working on them a practice in futility.

With an ICE, I get gas once a week or so. With an EV, I plug my car in overnight a couple times a week, maybe using a fast charger once a month or so when I realize I've neglected to do before a longer trip. With a hybrid, I'm plugging in every night, plus getting gas once a month or so.

With a (manual transmission) ICE, I get to decide my power curve, and know if I slam the peddle to the metal, I can get a LOT of power out of that car. With an ICE, my acceleration is limited mostly by software. With a hybrid, if your EV mode battery gets depleted, the car gets sluggish. Plus, you're dragging around a bunch of dead weight in the form of a battery, so you don't even get respectable acceleration on ICE standards.

On top of all that, hybrids cost more (because they have basically all the parts of an ICE and an EV).

I wonder if you separate plug-in hybrids and non-plug-in hybrids, whether you might gain a more positive opinion of some.

With non-plug-in hybrids, you don't plug them in at night, and you're lugging a lot less battery around. In some sense, you do have all the complications of both, but as an example, the whole planetary gearset, dual motor setup in Toyota's Synergy drive replaces a traditional gearbox and seems to be more reliable and more efficient. I also expected Priuses to have worse reliability than non-hybrid Toyotas, and seem to have been completely wrong.

Also, while what I wrote above probably makes it sounds like I'm against plug-in hybrids, I think of them as a way to reduce the weight of a full electric car, by replacing a lot of the battery weight with a traditional ICE drivetrain, i.e. a range extender/light-weight source of power for acceleration.

I also don't like the weight and complication of modern cars. It's really hard to beat ICE for weight, but seems to be pretty easy to beat it for acceleration and fuel economy at the cost of making it more complicated (and worse handling).

I read a review of the BMW 330e iPerformance once, which messed with my head. They made the point that BMW had basically found a way to make a car heavier, without hurting acceleration or fuel economy (and also not really improving either).

FYI: I am just interested in this area, and have spent way too much time thinking about it. Many people will be more knowledgeable in this space. I just wanted to throw out some ideas to be shot down :)

Edit: Forgot to comment on your Integra - probably nothing modern will rival that for a long time! I was crushed giving up my '94 Celica a couple of years ago when I moved overseas. Basically gave it away!

>The seller sets the price. Not the buyer. You should ask why the seller is willing to let the vehicle go for a lower than expected price for a given number of miles.

The seller sets the asking price; the buyer sets the bid price; and mutual agreement between the buyer and the seller sets the selling price. When sellers set an asking price, they tend to refer to information about previous selling prices, creating a dynamic where buyers influence asking prices.

>Do those often persist through private party sales?

Yes: "Your New Vehicle Limited Warranty will follow your vehicle and be transferred to the new owner when a vehicle ownership transfer is performed through Tesla." (https://www.tesla.com/support/vehicle-warranty) You opened your comment with "Why guess?", so I wonder why you posed this as a question (in such a way as to hint that the answer is "no") when the correct answer was a Google search away.

> but hybrids are still obviously better

Like a lot of anti-EV commentary^, there's nuance being missed and that is: use case. There is no "X is obviously better" (this is also true for a lot more topics than this)

I have a Nissan Leaf that's perfect for the purposes we bought it, better than a hybrid, far better than a petrol or diesel car. There are also scenarios where a hybrid is better, and there are also scenarios where a great big diesel burning engine is the best option. These gaps are closing though.

^I've had frustrating conversations with relatives when discussing the possibility of buying an EV where they just round-trip through a list of negatives, and I have to explain repeatedly that they don't apply to our use case. Examples:

- Can't tow: I haven't towed anything with any of our ICE cars in years (and we still have the ICE cars anyway)

- Can't take it interstate: We're not planning to take it interstate, and we're not planning on buying one big enough to carry what we'd need for an interstate trip anyway.

- Wouldn't a hybrid be better / safer first option to dip into electric: Not for our very regular and frequent short trips, pure electric is simpler and cheaper, and I've done the research to be confident in the decision.

- Miscellaneous other: We're not actually getting rid of the existing cars (yet) so we're not losing anything you might be worried about us losing.

The Leaf is on track to have saved us nearly $2k in petrol for the year (even taking into account the cost of electricity to charge her). If she lasts another four years she'll have paid for herself.

I agree with you. Cars seem to be discussed as an all or nothing decision process.

Many families where I am have two cars. Both cars spend an inordinate amount of time driving around the city. One car will be the preferred long distance family hauler.

Simply making the other car full electric seems like a no brainer to me.

In my family we kept the Impreza, but the electric (Tesla 3) became the long-distance hauler. It turns out to be cheaper to run long distances, quieter and more comfortable, and surprisingly has more room for baggage. We couldn't find a single reason to keep using the Impreza for long distance.

Charging options are not so well built out in my country. It'll get better, but it's not there yet.

So ICE for the long hauls, electric for the city running where you can reliably charge it at home.

Makes sense. My country, Israel, is barely 500 km from top to bottom. Even for what we consider long distances here, the charge from home gets us anywhere, and another charge gets us back.

  > I'm not trying to be mean, I think EVs are great, but hybrids are still obviously better

I was 100% with you up until this point. Hybrids have all the mechanical components of ICEs, thus need maintenance and repair just like an ICE vehicle. Additionally they have all the components of an electric. The reduced maintenance of EVS is their greatest strength, in my opinion as an EV owner who was not expecting this benefit.

I've replaced more transmissions and rebuilt more engines than most people you'll meet. The EV drivetrain, even if it has a weak point in the battery, it's such a pleasure to own because of the reduced maintenance. All the money saved on oil, oil filters, plugs, mufflers, replacement stolen catalytic converters, coils, plug wires, fire and air filters, valve gaskets, water pumps, timing chains, lifters, brake pads, and the damned automatic transmission, add up to about the same order as a battery replacement - if that battery replacement is even needed. In the ICE vehicle, all those things are definitely needed.

I completely agree. My father was a truck mechanic and I spent plenty of time under cars turning wrenches. I'm very happy that part of my life is behind me as an BEV owner. My wife still drives an old Accord and I dread the times when it needs an oil/transmission fluid change, muffler replacement, water pump, etc. All of that simply disappears when you switch to a BEV and it really is refreshing not having to think about it at all.

> Why guess?

Manufacturers DO publish reports. Tesla routinely provides data about the lifetimes of their vehicle batteries, which turn out to greatly exceed their warranties. (This shouldn't be surprising to anyone. Manufacturers just can't afford to have high warranty return rates).

The rest of your post is an attempt at hand-waving away what is a real phenomenon. EVs are still new to a lot of people, there's a lot of FUD floating around, and that can affect used vehicle prices. Markets behave irrationally sometimes, and that leads to opportunity. (In this case, getting a value buy on a used EV).

> hybrids are still obviously better

Hybrids are objectively worse. They have less all-electric (efficient, non-polluting) range, they charge more slowly, they have a more complex drivetrain with more parts subject to repair, etc. They're a better fit for a specific market segment, but that segment is small and rapidly becoming non-existent. You can get cheap EVs nowadays that have more range than my last ICE.

> that segment is small and rapidly becoming non-existent

Disagree. People vastly overestimate the complexity of hybrids.

The mechanical and electrical components in a Toyota-style planetary gear based hybrid are much simpler than a standard automatic transmission, and demonstrably more reliable than both a conventional automatic and belt-style CVT.

This is a long way of saying, the specific market segment hybrids are a good fit for is the set of all passenger car customers that an EV is _not_ a good fit for. Anyone buying a Corolla, Civic Rav4 or CR-V should be buying the hybrid, and sales seem to be trending that way.

[Hey I own one of those]. Nah, the majority of people buying those specific vehicles should be buying an EV instead. In fact, this article is targeted specifically to them.

Rather, the biggest need for hybrids are people who frequently tow long distance at highway speeds. The combined aero drag is ridiculous: about 25% of typical EV efficiency. You'd need to drag a humongous 300kwh battery around to get range comparable to a typical sedan while towing.

Instead, some of the new EREVs are more like a BEV + hybrid, in that they have a BEV-sized battery (e.g. 100+kwh) for all your non-towing driving, along with a generator to handle the long distance towing.

> Tesla routinely provides data about the lifetimes of their vehicle batteries,

They were caught lying many times.

I just traded in a 6.5 year old model 3 with 75k miles.

Battery was at 87% of capacity.

The big problem was cold snaps. It had the older heating system and would lose a lot of charge in the cold. Our 2022 Model Y with the newer heating system doesn't lose nearly as much charge in cold snaps.

Just had our PTC (resistive electric) heater replaced in a 2018 3 with 110k miles. Sure wish it had a heat pump, but we don't suffer much range as we live in a mild climate.

does the newer one use a heat pump?

Oh yes, the entire heating/cooling system is quite a beautiful bit of engineering. a very elegantly designed “supper manifold” and heat exchanges that can push or pull heat from any device to another in the vehicle. They don’t even have heating elements anymore, they just run the motor less efficiently to produce more heat!

Yes, they do. All the newer teslas use heat pumps (as well as most other EVs these days).

A lot of them it’s still optional which says a lot about their small scale.

Yeah, the old one heated the battery through similar tech as heated seats. I think the only problem with the tesla heat pump is that it doesn't function well below 20-30F. I'm not sure if that's all heat pumps.

Well, the problem with the heat pumps in my house, is that near or below freezing temperatures, the outdoor unit has the tendency to freeze over, then either it has to run its resistive heater, or intermittently cut out and blow some cold air inside. It's rather uncomfortable.

I'm not sure you can get away with a design that has to sporadically turn itself off, and melt itself down in an EV.

Something is wrong with your heat pump. Mine works fine during cold snaps. It doesn't blow cold air during the defrost cycle.

Perhaps yours is undersized, or otherwise not designed for cold climates?

Heat pumps on cars can go well below freezing without any issues.

> Most manufacturers offer a warranty somewhere in the range of 8 to 10 years, and 100,000 miles. That usually means that if your battery is below 70% health within either 8 years or 100,000 miles, they’ll replace it for you.

This disagrees with pretty much every other source I've been able to find. An N year/M mile warranty is good until whichever of N years or M miles you hit first.

The quote you reference doesn't make a distinction either way.

Doesn't it? It says:

> That usually means that if your battery is below 70% health within either 8 years or 100,000 miles, they’ll replace it for you.

Consider a car with 200k miles at 5 years. 5 years is within 8 years, so I don't see any way of reading "within either 8 years or 100k miles" that would not make that car covered. Similar for a car that is 15 years old with 50k miles.

But everything I've seen elsewhere says the warranty applies as long as you are within 8 years and within 100k miles.

Sorry, I was parsing the "either" incorrectly here.

Well played Sir.

My MG4 (Saic) has a five year warranty and rather a lot of small print, that I should probably OCR and pass through Chat for a summary and then read myself.

I do know it will last longer than my last car - a Ford Focus with an Eco boost abomination of an engine. The engine oil warning gauge thing decided to only kick in at just above the bottom level on the dipstick. Even so all should have been fine. I even drained and replaced the lot. This was a three year old car. It degraded over about a month and eventually passed away in Bristol. I live in Yeovil, a good hour away.

My first car was an elderly Mk1 Ford Fiesta (it was old in 1994). That often managed to run without any oil for some time because it had ended up as a rusty stain with the contents of the radiator on the underside of the bonnet (hood) and on the road.

I agree. and besides, that warranty is way too short. after 8 years, the car is unsellable with a 60% battery. And the used market is trending that way

It's stated directly in the article that Tesla's study reports vehicles with 200K miles generally retaining 85% capacity. That's nearly 300% better than what you're suggesting.

first off that's not the case just check everyone running their battery calibration tests on forums.

And the context above is the warranty not covering long enough durations. 70% or 8 years does not protect the value of the vehicle. Below 70% @ 9 years is a worthless car

> first off that's not the case

It's objectively the case. They literally have the data to back it up, regardless of what anecdata you might have seen on some forums.

> And the context above

Their fleet data says 85% battery average capacity at 200K mi and that battery degradation slows down over time. That's a far cry from 70% at 9 years.

Not to mention what’s the stddev ? That’s still a lot of write offs even going by their subjective figures

That’s the opposite of objective.

> first off that's not the case just check everyone running their battery calibration tests on forums.

Relying on posts from forums is going to end up with some major selection bias. People who don't have significant battery degradation are less likely to go to forums to see what others have been dealing with.

i agree partially. I think both outliers will post. the good ones for bragging rights.

But don't be too quick to dismiss. there's an entire industry in tesla battery repair to address the loss in value. That wouldn't exist if the batteries were at the claimed level.

and there are many other indications.

It's funny how "statistics" from tesla will be taken at face value, whereas other real experiences from owners are dismissed as "anecdotal".

They are all just stories. Who do you want to listen to?

I think we (sorry I) have seen that degradation has not the concern, it's the pack engineering that is an issue by a large margin.

Tesla's packs first produced in 2017/18 for the model 3 represented largely the industry's first mass produced packs that will largely fail naturally, not due to pack engineering issues (failed cells, leaks, cooling, etc...). Before that required a much higher pack replacement rate, and other manufacturers have the same issues.

Also the early Nissan Leafs, pioneers in the mass-market EV space, had batteries with only air cooling and which experienced significant degradation.

More modern EVs with full liquid thermal management and newer cell revisions and chemistries seem to be holding up much better over time.

Some chemistries like LFP have even greater cycle and calendar life in return for a bit less energy density. Ford and GM are both betting big on these for their future entry-level EVs and I think they will end up being a common choice where maximum range isn't the customer's primary concern.

>Also the early Nissan Leafs, pioneers in the mass-market EV space, had batteries with only air cooling and which experienced significant degradation.

Don't forget that beside the chemistry issue in hot environments, the original Leaf only had a 24 kWh battery, so you'd have a lot more cycles than say a 60 kWh or 90 kWh battery. If you assume it is good for 1,000 equivalent charge cycles, and assume you 3.5 miles/kWh, than your 24 kWh battery would be good for 84,000 miles. A 60 kWh pack would be good for 210,000 miles, and a 90 kWh pack is good for 315,000 miles. A new Chevrolet Silverado EV has a 200 kWh pack (which, if you can squeeze out 2 miles/kWh, would be good for 400,000 miles).

And with a small battery it is more likely that you'd need to charge up to 100% and discharge closer to 0%, which is also harder on the battery.

>>Also the early Nissan Leafs, pioneers in the mass-market EV space, had batteries with only air cooling and which experienced significant degradation.

Volkswagen e-Ups, Seat Mii Electric and Skoda CitiGo EV(same car really), all have an air-cooled battery, been on the market for 13 years now and there's no significant degradation reported(not in a systematic way like with the Leafs). I think it's just different chemistry to what the Leafs were using.

How do those compare to Leafs driven in the EU? Leafs in the US market certainly experienced large degradation. None of those other vehicles have been available in the US market.

Wondering if the driving pattern of US vs EU drivers or the more extreme temps (high and low) in the US are the real culprit in the degradation of air-cooled batteries.

EU leafs have also massively degraded, the first gen leafs go very very cheap because you can do like 40-50 miles on the battery at most.

> Ford and GM are both betting big on these for their future entry-level EVs and I think they will end up being a common choice where maximum range isn't the customer's primary concern.

A lot of people think they need way more range than they actually do, especially people that have decent charging at home. The think they need long range for the occasional long road trip but even there range is less important than they think it is. For long road trips charging speed is more important.

Briefly, consider a 3000 mile road trip. If your average highway speed is 75 mph that's 40 hours of driving. On top of that 40 hours you have whatever stoppage time there is to refuel/recharge. Let's put a lower bound on that.

Suppose you are in a car that can go 200 miles between stops. When you reach the first stop there are 2800 miles left, so over the course of all future stops on the trip you have to add a total of 2800 miles worth of fuel/charge. Let's say your car is an EV that can add 1000 miles in an hour of charging. That's 2.8 hours of total charging time for the trip.

Someone else in a 300 miles EV but that only adds 300 miles in an hour will first stop with 2700 miles left. They have to add 2700 miles of charge over the course of the trip, which takes 9 hours.

They will have fewer total stops (9) than the 200 mile range fast charging car (14) which favors the long range car because each stop has some overhead that is not spent actually charging but that is unlikely to be enough to make up for the slower charging car spending 6.2 more total hours actually charging.

Where the 300 mile slower charging car shines compared to the 200 mile fast charging car would be for people who have frequent trips like Los Angeles to San Diego and back. The 300 car could do that on one charge. The 200 mile car would would need a charge stop.

Similar for people who have trips like Los Angeles to Los Vegas where they will stay overnight. The 300 mile car could do that with about 10% to spare and then charge at your hotel. The 200 mile car would need a charge stop before reaching Vegas, then you should charge at the hotel, and then you will need one other on the way back. (If you charge to full on the first charge stop you can skip the one at the hotel but then the one on the way back will be longer, so you are better off taking the 3 charge approach that includes the hotel since that one can take place while you are busy losing your money in the casino).

For what it's worth, longer range EVs also tend to support faster charging.

A 250 kW Supercharger only charges at the full 250 kW when the battery is under I think around 25%. After that, it falls off linearly as the battery fills. (EDIT: This is why it's better to do somewhat more frequent, but shorter charging stops. On my road trips, I typically charge to just 60-80% with the expectation of arriving at the next charger with 10-20%)

So if you have a 100 kWh battery compared to a 75 kWh battery, the 100 will give you more miles per hour of charging, simply because it can handle the full 250 kW charge speed longer.

Sure, but the battery types you list (lower capacity with very fast charging vs higher capacity with slower charging) don’t match what is actually in the market.

The best way to get a vehicle that charges quickly in miles-per-hour is to get an efficient EV with a large range and just charge it in the fastest part of its charging curve, usually around 10-60%.

The best cars on that sort of roadtrip metric currently are the Porsche Taycan and Lucid Air because they have large, fast-charging batteries in an efficient form factor.

See https://outofspecstudios.com/10-challenge

>Also the early Nissan Leafs, pioneers in the mass-market EV space, had batteries with only air cooling and which experienced significant degradation.

I never understood why they carried on with that for so long when they had a better system in their vans which they could have dropped right in.

To this day, I think the existence of the first-gen Leaf was actually harmful for EV adoption.

It had the trifecta of short range, slow charging, and nasty degradation that all feed into the negative perceptions of EVs.

You may be right. But we have a Model S 85D from 2015 and basically everything was replaced (seats, all door handles, ac compressor, sun roof, glove box, gauge cluster LCD, main LCD, MPU) except the battery. That's been great, and 10 years in tracking at 85% capacity.

> But not many cars get to this driving distance

That's just because they don't receive appropriate maintenance. In my family we had plenty of Italian and german cars, we maintained them, most hit 300k+ kilometers. Our 9000$ Lancia Y still worked fine after 350k+ and we only got rid of it because it cannot enter Rome due to emission restrictions.

Italian cars work great in the warm and dry Italian climate, but have historically had trouble with corrosion in colder climates that they were not built for. My dad loved Alfa Romeo’s, but none of them lasted very long in Denmark. In other words YMMV.

We had an Alfa Romeo and it did not enjoy -20°C. That model has almost completely disappeared from the market after 20 years, with the remaining ones usually being sold as projects or for parts.

I think they also had problems with timing belts? Google results are suggesting me that they had to halve the change interval, possibly because of our shitty roads. Volvo belts also last for 10 years in their native Sweden but only 5 years here.

Italian cars from the sixties through to the eighties were notorious for corrosion. It did mean you could buy something interesting for not very much money though. After the nineties they got a lot better.

https://xkcd.com/3123/

I believe the phrase was originally about "mileage" as in fuel economy (i.e. miles per gallon), but "total mileage" (i.e. odometer reading) is pretty close!

True! Nice pun, jacobgorm :)

Please don't do this, you aren't adding anything to the conversation

I mean, you can get any car to last forever, it's just a question of being economically reasonable. My mum still has a 2004 Land Rover Discovery 3, one of the original ones with a 4.4L V8, and that car has a half a million kilometers(300k miles) on the clock now. Everything works, inside and out, I drove it at motorway speeds last month and it still felt super stable. And these cars have a terrible reputation for essentially ruining their owners and the electrical systems going haywire. The secret? My mum spending an equivalent of anywhere between 2000-5000 euro a year(!!!) on servicing and repairs to keep it in tip top shape. The car is probably worth only 5k, maybe 8k at a push. It doesn't make any logical sense, but it can be done.

Maintaining a fiat car in Italy is very cheap.

Wow, I did not expect anyone to be offering a SIX HUNDRED THOUSAND mile warranty on their batteries. That's some serious confidence. I didn't see anything about it transferring, though. That would be smart on their end - the resale value for electric sports cars at least, is about 50% in the first year, then it levels off hard after that. This would encourage buying new, but not aftermarket. I'll have to look into this.

Still, while this removes a primary concern of mine, there's still one major hurdle that cannot be bypassed as far as I can tell (yet): If you have shared parking, there's essentially no way to charge your car. Maybe if it's an outdoor parking lot you can rely on solar power somewhat, assuming you're in a good situation for that?

Still, my point is that my parking space isn't actually mine, so I can't modify anything in the garage. Assuming superconductors aren't figured out any time soon, this appears to be an impossible solve, which cuts their consumer market significantly.

Also, not exactly the same thing, but they could remove those warranties and instead get some nice replaceable battery cells in there. Let me turn a thing to unlock it, pull out that one cell, and replace it. But maybe I'm a little more wrench-y than their customers want to be?

At my last apartment before I moved into a home where I did have the ability to install a charger, they had 4 EV chargering spots in the parking garage. I believe residents just had to pay the normal residential electricity rate to use them, they were standard commercial level 2 chargers like the kind you see in public parking lots.

All this to say, if the demand is there then shared parking structures will install them. I live in a city with a fairly high percentage of EVs, but it will continue to spread.

We get away with level 1 chargers, and live far from the city. Residential lots could easily get away with one level one charger per spot. (The wattage is < 25% that of one level 2 charger, so you can put in 4x as many with the same backend connection to the grid.)

For city commuters, this would probably be more than good enough.

Yeah, Level 1 charging is way too easily overlooked in the US. A lot of US parking lots could add simple Level 1 outlets to most lamp posts and do a lot, easily, for EV charging. (Most traditional halogen lamps were nearly Level 2 circuits, prior to recent switches to LEDs. If the LED transition had been timed a little different there might be way more L2 chargers "easily" installed in parking lots.)

A bit of an aside: I think part of the public perception problem is calling Level 1 "chargers" and not just "outlets". At so many points in our discourse, especially in the US, we've let car manufacturers sell us this idea of "gas-pump-like capital-C Charger" as something bulky and "hard/expensive to install", but really most EVs just need more wall outlets, classic, boring electrical outlets. Sure, the US can blame Edison that we don't have Level 2 as a default outlet and our cheapest/easiest outlets are Level 1, yet still we need to stop underestimating L1.

The other thing beyond "don't discount L1 as a reliable way to charge" (slow and steady charges the race car, eh) is "don't discount the power of destination chargers". Everywhere you park is a possible place for a charger. If you can't get one easily at home, maybe your employer can build one. Your grocery store and your church or bar or pickle ball court or other third place can build one. (Especially Level 1. Outdoor outlets have always been a thing, moving them a little closer to parking spaces shouldn't always be a big deal. Boring old electrical outlets are "everywhere" already, we just aren't always yet in the mode of thinking about them, their ubiquity, and how they can charge our cars, while we eat or shop or work or hang out or play or sleep.)

I try to make this point all the time.

We have already built all of the charging infrastructure except the chargers.

You are probably no more than a block from a 440v line (that's what typically goes into the transformer, of which there is one on almost every block, at least). You are in a building that almost certainly has 220v power in it. You are probably less than 10 feet from a 15-20 amp 110v plug for almost your entire day. There are far more places to charge a car than there are people in most of North America.

If the incentives are correctly aligned, we have the infrastructure to make this happen VERY quickly.

Electricity generation is an issue, but not as much of one when you realize that not every car will be charging at the same time. Not every car will be fast charging. Hell, not every car will even need a full charge every week. I fill up my ICE car every 400km or so, which is about 2-4 weeks depending on weather. Right now AI growth is projected to increase the rate of power consumption far more than electric cars even under the most optimistic adoption curves. If generation is the problem, we need to kneecap chatGPT immediately.

Its amazing how many people think that our gasoline infrastructure is a given, and that electric car infrastructure is impossible.

A related tale: One of Rivian’s goals was to electrify National Parks and campgrounds as part of their branding as a “rugged” or “off-roading” brand and one of the maybe funniest things about that was how unsuccessful they were in parts of that branding effort, not because it was hugely expensive to get electricity out to such places but as much that it was hugely silly to take credit for all the electricity already there.

Any US park or campground with regular RV visits has tons of 240v “dryer plug” outlets, many of which with decades of battle testing of simultaneous high draw use. EVs look almost considerate next to most RVs, and those don’t even use any of that electricity to drive.

SAE J1772 (the ugly “gas pump” looking thing that CCS in Europe still resembles, but the US is fortunately moving to the relatively saner NACS instead) should have just been a “dryer plug” and we might have avoided some silliness in how few people realize how much existing EV infrastructure exists and/or what can be repurposed easily as EV infrastructure.

I think the biggest hurdle to just doing that is who pays for the electricity. Sure right now it's a nice perk you can provide your EV owning visitors that probably won't cost too much, but in a world where 10%+ of cars are EVs the costs will add up even at level 1, so you'll need to go for capital-C Chargers that come with payment infrastructure.

one answer would be to make the parking spaces expensive enough to account for electricity. parking meters are pretty widely adopted.

Arguably a lot of parking fees are already expensive enough to account for electricity. In a city charging, say, $15/hour for parking, it seem like we should be able to expect say $0.15/hour to be a drop within the existing profit margin. There's a very simple order of magnitude relationship between the two numbers already.

WiFi-controlled 120V outlet plugs cost $20 or less retail (including tariff costs.) Those aren’t rated for the sorts of continuous draw an L1 charger needs, but upgrading the hardware to handle this isn’t going to make the hardware crushingly expensive. So the actual question mostly comes down to software and integration. Seems like a good ycombinator business. Think of the TAM!

I do keep joking that I've got hundreds of dollars to invest in the first restaurant that wants to be the McDonald's of EV Charging. (The early business model of McDonald's was notoriously putting one near almost every interstate exit so it became a ubiquitous staple of the driving landscape.) Could be McDonald's itself. The Sonic-style of sit-in-your car "drive in" restaurant seems almost like a preternatural echo of an EV restaurant business. It would be a beautiful irony if Cracker Barrel decided EV charging was the next great idea; sit in a rocking chair and peruse the gift shop of very analog goods while you wait for your car to charge sounds like a smart business model to me.

Some company that wants a restaurant near every interstate exit to build their brand is going to figure out that the economics of electric charging are simpler than they expect and with it they build a potentially very loyal audience of travelers with easily 30+ minutes a stop on their hands to eat, shop, what-have-you. Maybe it will be one of the old guard of such restaurants, maybe it will be a new disruptor. If someone on HN wants to start it, I have a very tiny seed fund round in the hundreds of dollars to invest.

Yep absolutely, I used a level 1 charger at home for a couple years and it could easily recharge my daily work commute in about 5-8 hours (depending on season). Even now the only upgrade I did was move to a 240V16A charger because I wanted it to be a little quicker after long trips, but most of the time I limit the charge rate to 8A to preserve battery health.

There are some cars with panels, but they can only get about 10ish miles a day with good sunshine. Stationary panels work much, much better.

> Assuming superconductors aren't figured out any time soon, this appears to be an impossible solve, which cuts their consumer market significantly.

What does that have to do with EVs? The inflection point for adoption is solid state batteries, and there are some experimental factories under construction. (Solid state batteries don't loose charge when parked and can charge about as fast as filling a tank of gas.)

> Also, not exactly the same thing, but they could remove those warranties and instead get some nice replaceable battery cells in there.

Battery exchanges are impractical because the battery is part of the frame.

I don't think the bottleneck for charging is in the batteries, it's in electricity as an energy vector. By its very nature, someone is either instantaneously dispatching it from somewhere, or it's already being generated and curtailed. I just don't see that being cheaper than even biofuels in the long run, because time arbitrage matters. Making fuel with overcapacity that is worth zero (or less!) probably scales better than trying to store it all in batteries, because holes and containers will always be cheaper and easier to expand.

> Solid state batteries don't loose charge when parked

citation needed. /s

Never heard of batteries without self-discharge.

> If you have shared parking, there's essentially no way to charge your car.

The neighbourhood I used to live in London (where almost nobody has off-street parking) installed chargers into lamp posts. This BBC article has more details and photos https://www.bbc.co.uk/news/business-67518869

Home charging in shared parking scenarios is difficult. Municipalities can add curbside chargers and in some places this is fairly common. In a private condo or apartment scenario you'd need the owner or association to agree to install them.

A second option is more slow chargers installed places your car spends a lot of time parked, like offices or transit stations if you park and ride.

A third option is using a fast charger somewhere you go once or twice a week. Like grocery stores, gyms, etc. Costco for example is adding fast chargers to their stores, which should be fast enough for a full charge by the time you actually get in and out of Costco.

Replacing cells in a pack can be difficult. You want all the cells in the pack to have roughly the same capacity and voltage curve, so that you can connect them all together and charge them at the same time.

GM says that their Ultium batteries are segmented into modules, which each module having its own Battery Management System, and that it supports mixing and matching modules of different degradation and even cell chemistry.

But anything that adds complexity to the pack beyond being cells packed in as densely as feasible is going to add costs and reduce maximum energy storage.

I think the long term answer here is that there will eventually be a used and remanufactured battery pack market for popular models, just like you can get a used or remanufactured engine today.

> A second option is more slow chargers installed places your car spends a lot of time parked, like offices or transit stations if you park and ride.

I don't think this will ever happen. It's the worst case in most every sense. You're talking thousands of chargers, for most parking structures, to solve a problem that's mostly about current battery tech/infrastructure. When battery tech is ready for general use, this won't be needed.

Battery tech is for general use. The median and mean usage of a car in the US is 40 miles per day. A 300 mile battery gets you a week's worth of driving between charges (~7.5 days). That's comparable to a median ICE car that gets 300 miles on a tank, with the subtle distinction of needing a 30-45 minute fast charge versus a 5-10 minute refill. But that's still a once a week "problem" with useful mitigations such as it is dangerous and illegal in most states (just poorly enforced in many as well) to leave a car unattended while refueling with gas, but electricity is far safer and multitasking is easier and more convenient while fast charging. (That fact that most fast chargers aren't interesting destinations with enough things walkably nearby is a different problem to solve, that the market should be rather good at solving eventually.)

But that's all still treating EV charging in the old world ICE model which everyone is familiar. When people are talking about wanting more chargers everywhere a car may be parked, like offices or transit stations and other parking structures, that isn't a need, that's a market opportunity unavailable to ICE. You can't put a gas pump in every parking space, but you sure can put an ordinary electric outlet. We can distribute the charging "problem" of a car far more easily than the current centralizing forces of gas logistics. It's an amenity that anyone who owns a parking lot or garage can offer to encourage walkability to nearby businesses and/or homes. It's a possible revenue source for other parking lots or garages that love low margin business models like electricity metering and/or think they have a captive enough audience to charge whatever margins they like, to make the bottom line grow.

We don't need those things to happen. We've driven gas engines for enough decades without that. We want those things to happen. We expect market forces to eventually deliver those things, as soon as the market better figures out what EV charging disrupts in parking lot planning and operations/maintenance. You can't expect your gas car to have more gas when you come back to it in a parking lot, but an EV can have a slightly higher charge almost anywhere it is parked for a while and that's a game changer that will slowly spread as the market finds the fun (and profit or marketing opportunity) in it.

Battery tech is ready for general use. Over 20% of cars being sold in California are EVs now, and over 90% in Norway.

Slow chargers are pretty low-tech devices, just a 208V-277V circuit with a device that handles switching, ground fault check, and potentially payment. These are going to be cheaper and easier to install and maintain than fast chargers, and I think adding them to workplaces is going to be easier than covering individual apartments.

That certainly won't cover all needs, which is why I listed other alternatives as well. The answer will be a blend of these solutions where each makes sense.

> Battery tech is ready for general use.

> Over 20% of cars being sold in California are EVs now

These are not compatible, if you're talking BEV. Regardless, you've provided data showing that it's not acceptable for the overwhelming majority of buyers, which matches market research [1]. And, out of that group, 30% want to switch back to gas [2]. Cost, and the massive depreciation is a factor related to current batter tech.

And, what % of commercial vehicles sales are BEV (which is included in general use)?

> and over 90% in Norway.

In California, 2023: 25% sales. 2024: 25.3% sales. There trend has slowed, maybe related to our ridiculous electric prices (fuel is cheaper, in many cases). It's very difficult to compare small European countries to a something in the US.

For some anecdotal evidence (which seems somewhat sufficient for the definition of "general"), I own an EV, and I know it's not ready for general use, because I will not be selling my gas car. In fact, I'm replacing it next month with another gas car. Most people I know have an EV (like 70%), and the majority have a second gas car that they say they will not sell. The majority of those that only have an EV say their next purchase will be a hybrid, all matching the trends shown in market research.

I'm crossing my fingers for another salt battery breakthrough, which are making their way into BEV [3].

[1] https://www.mckinsey.com/features/mckinsey-center-for-future...

[2] https://www.cnbc.com/2024/07/25/ev-owners-want-to-buy-gas-ca...

[3] https://www.reuters.com/technology/chinese-battery-maker-cat...

I don’t think it’s the battery specs holding back most buyers. For one thing transitions take time even when new tech could cover most of the market’s needs. But the main thing currently is purchase cost and charging infrastructure availability, both of which are improving at a fairly steady rate.

Most of the work going into scaling up EV production currently is about producing higher volumes of the batteries we have to bring costs down. A second prong is working on higher energy density and faster charging, but these solid-state batteries are going to be expensive and start in high-end vehicles, not economy cars for the masses.

> I don’t think it’s the battery specs holding back most buyers.

First link shows it's, at least, the primary (majority) concern for purchase. Price is part of the battery tech/chemistry package (~40% BEV price is battery), which seems to be the real killer:

> Only 33 percent of global survey respondents say they are likely or very likely to buy an EV at current price levels

Yes. But scaling up current battery tech is what the industry is doing, and that will bring down prices.

Improved charging infrastructure also offsets range concerns.

> talking thousands of chargers, for most parking structures

My home has an average of 10 chargers per room; I don't think it's really been a big driver of its cost.

I imagine the chargers you have are not drawing 3kW each though.

That's the main problem - your legacy infrastructure is most likely wired for 220V@32amps for the whole garage/street just to run the lamps from it, so 7.2kW. That's one EV charger, or two if you want to split them into 3.6kW feeds. If you want to run a proper 7.2kW charger from every lamp post or next to every parking space, that's a lot of brand new cabling that you need to add.

Yes, but potentially easier than adding 250kW per charger for a bank of DC fast chargers.

The grid connection for one fast charger could serve 50+ L2 chargers, potentially even 100 with load-sharing chargers.

There are good use cases for both kinds.

1.6kW is the limit; but no, they aren't. But you don't need 7.2kW all the time! There's no way that every single car would need to charge at every moment, and I know this from walking through parking garages and seeing some cars not move for days at a time.

A EVSE could easily serve multiple spots, and fairly (or unfairly, for profit!) distribute power between cars from a limited supply

Please note, the context here is level 1 charging. 7.2KW is level 2.

With level 1 charging is only 3 to 7 miles per hour, so average of 35 miles in a 7 hour day (assuming you drive for your lunch break). Where I am, the average distance to work is around 27 miles (one way), so a net loss of charge.

But the reason you drive such a long distance to work is to be able to live in a suburb, right?

Looking at the actual stats, you're a bit of an outlier: https://aaafoundation.org/wp-content/uploads/2023/09/202309_... (p. 12). The average commute trip is 22 miles, and keep in mind you also could use a level 1 charger at home.

No, it's 22 miles to/from work, one way [1]! My commute distance is only a few miles more, and my commute time is almost exactly the average time listed there.

For most, the purpose of living further from work is reduced total cost of living, especially if you're near a big city [2], where it's not usually an option. I save thousands a month by commuting a little, for the same number of bedrooms (which has a legal minimum where I am). If I wanted the same square footage, I'm saving > $10k/month, compared to being 1/3 the distance from work.

[1] https://www.census.gov/topics/employment/commuting/guidance/...

[2] https://www.tandfonline.com/doi/full/10.1080/02723638.2022.2...

This is already being done in Finland.

Shopping malls have parking with literal hundreds of chargers.

Even my local grocery store has 6 Level 2 chargers and 4 level 1 chargers.

Hell, the McDonald's has four 350kW chargers. Modern cars charge faster than you can order a Big Mac and shove it down your throat =)

The solution is to have Level 2 and Level 3 charging _everywhere_. Then you don't need to have massive 100kW+ batteries in cars, because you can get a bit of a charge every time you park.

The first big leap people moving from ICE to EV have to get over is that unlike ICE cars, EVs don't "need" to be charged to full every time. You plug in, do your stuff, unplug and move on. Very rarely does one need to "go charge" during normal daily life, that's only for road trips.

> The solution is to have Level 2 and Level 3 charging _everywhere_.

I somewhat agree with this. My response was specifically, which I quoted, about "slow chargers" at office or transit, which I assumed was level 1. Level 2 is 8x faster, and can accommodate a 25 mile commute in two hours, rather than a net loss of change over 8 hours.

Level 1, at these locations, doesn't make sense.

Depending on the level 1 though.

My "level 1" at home charges at 230V/10A from a normal outdoor outlet, which is around 2.3kW/hour. During an assumed 8 hour work day that's about 18kWh, which is enough to move my EV for over 100km.

My commute is WAAAY less than 100km, meaning that the battery is, in practice, full every time I leave. Even dropping it to 8A would still be perfectly enough.

The biggest problem is that L1 chargers are just glorified wall warts, nobody is doing L1 speeds with proper integrated Type2 connectors.

Do you believe mass deployment of level 1 is going to happen?

It already has, Finland has "level 1" at pretty much every parking location for block heaters.

All we'd need to do is remove the 2 hour hardware timer from it and make them 24/7.

> Still, my point is that my parking space isn't actually mine, so I can't modify anything in the garage.

Presumably over time shared parking areas will get upgraded with charging infrastructure to keep attracting tenants.

The housing and rental markets currently favour owners/landlords significantly and it's not looking like slowing down. I have zero hope that "charging infrastructure" will be installed to "attract tenants".

Here in Australia landlords seem to struggle with basic things like insulation or a split system aircon.

> Assuming superconductors aren't figured out any time soon, this appears to be an impossible solve, which cuts their consumer market significantly.

I don't think superconductors solve anything in the EV charging space, and certainly wouldn't make L2/L1 charging easier to install for shared parking / street-side parking. An L2 charger uses something like a electric clothes dryer circuit, with 240V at 40A. Or somewhere in the 6-10 kW range, to recharge you overnight.

Room temperature super conductors would be huge for the EV charging space. One big thing holding back EV transport trucks is charging capacity. The infrastructure to support charging a fleet of EV trucks at 1MW+ each is insane with current tech.

Even supporting a bank of 350KW chargers is pretty nuts like you see in some highway rest stops (or Tesla Supercharger stations). Locations are limited by the proximity to high voltage transmission lines right now (ie: it's really expensive to push that much current for any significant distance).

If we had cheap and ubiquitous super conductors that could be run like regular ole medium voltage electrical cables....game changer.

One also needs good connectors.

There is also just the situations where cars are parked on the street and the cabling has to get across the public pavement to charge the car. Even though those people can deploy a charger they can't be blocking the pavement. There is a real concern here where the incentives for the individual to pay to deploy charging capabilities in their car parking bay or front garden can't actually do so because of ownership. It needs solving via legislation, a basic default that people can pay to deploy these systems themselves.

Charging on public infrastructure ought to get there in time but the really big benefit of electric cars comes when it charges at home on cheap electricity and the only time you worry about charging it at all is when you do a long trip and you have to charge it at the half way point for 30 minutes.

I live in Portland OR where electric cars are fairly popular. People just run an extension cable out to the street and put a cable cover over it on the sidewalk.

Where I live in Canada, that's illegal. Tripping hazard on the ground, I don't know the exact reason why overhead is also illegal (though I can make a few guesses).

I don’t know about the law, but the cable covers are the big industrial ones they use at concert venues etc. So no significant tripping hazard.

I know. Those cable covers are also illegal, something city law enforcement and spokespersons confirmed. Don't ask me why a cover is illegal, it seems strange to me. Maybe it's because of the danger of snow clearing equipment chewing it and the cable to pieces, but I'm just spit balling here.

And it's not like it's a new issue, I've heard public complaints about lack of solutions to this brought city council for years now. Haven't heard anything about the law being changed at this point.

It might be illegal here, too, for all I know. So is parking on the wrong side of the street, or blocking your own driveway. Some laws get ignored and no one cares.

Installation of AC Level 2 charging in garages is a technical problem but not exactly a problem on the level of “superconductors”. You need to install wiring and upgrade your service connection, and also install chargers that can share a circuit (which is commercially available.) It’s just a problem of figuring out who pays for it.

Certain ev sports cars are bought so that you get the privilege of buying nicer cars e.g. the taycan

That's true of any car, not just EVs. The word in certain circles is that people are only buying the Ferrari Purosangue in hopes of getting on the waiting list for their more desirable cers.

And the Taycan is a great car in its own right, just the price is stupid. The Audi E-Tron is the same car just packaged differently and can be had for sometimes half the price.

Electric owners under-value the last 20% of their battery. That is the most important 20%

If you are running out of "gas" , every lost mile is a mile walking (or being towed). that last 20% of range is the difference between making it to the next charger or being dead on the road. And with electric it's a bigger burden because they can't be refilled with gasoline.

As a practical example, my recent charging forecast dropped from 12% to zero % during the drive (this was controlled for consumption, ambient temps, driving speed etc). We finally arrived with 3% on the battery. So that means in a year, we will not be able to make that exact same drive. That is a problem needing addressed.

I've also not heard great things with the warranties. It seems people struggle to redeem compensation via warranty. And the qualifying conditions are not helpful for most customers experiencing poor performance.

I'm happy with my electric car, but I don't think more of the market will adopt them until this issue is directly addressed. "only 20%" dismisses the most critical and insecure experiences with the car.

Whether it's the first or last 20% of the battery is a software problem: People adjust their driving habits to avoid running out of charge, in the same way that people adjust their driving habits to not run out of gas. A car losing power when it claims to still have 20% left is a big problem, because it's failing to present you with the information with which to base your plan. If the readout on my car simply said that it had 240 miles of range on a full charge instead of 300, there's a good chance I wouldn't even notice.

On the exceptionally rare instance that I'm driving more than 200 miles in a day, I appreciate the half an hour to stretch my legs and grab a snack while my car's on a fast charger.

it's only software until the hardware becomes the boundary condition.

You're right for many drivers with dense charger coverage it's a nuisance.

But there are drivers who take trips with sparse charger coverage , where the 20% loss means insecure or impossible trips.

You really need to be in the middle of nowhere for a 200 mile range to be a problem. I've made quite a few trips in my Bolt where I worried there wouldn't be coverage (rural Indiana was my biggest concern), but it consistently turned out to be unfounded. In rural areas chargers certainly thin out, but you can safely drive the interstates and the vast majority of state highway with less than 100 miles between chargers.

The fact is you felt insecure because there was a risk. And with growing degradation means more of a buffer is needed. A 300 mi range vehicle is 192mi due to the buffer, and even that dwindles.

Only half of the non-Tesla EV chargers work, so your 200 mile charger now takes 6 hours to do the next 200 miles.

Even super chargers only function well over the bottom 50% of the battery . So the usable window is shrinking and shrinking.

All of these practical insecurities will need to be fixed before EVs expand beyond just a niche product.

You can dismiss them as silly or user error, but you won't sell any more vehicles that way.

Can you please share some (non-byzantine) trips that you can't make with only 80% battery capacity? Let's assume 240mi (80% of 300), since that seems to be about the average EV range nowadays.

Yeah, I had a model S for about 4 years and struggled with the bottom 20%. Everything becomes a stressful game of can I make it if the wind is too high. Part of this issue will be solved with more charging stations but the other part is degradation as you said.

I can relate. I've gotten pretty confident by watching the *consumption tab -- but i don't think most drivers should have to .

I think the ones who dismiss this issue drive regular routes with the EV and don't push the boundaries . Then there are the rest of us who actually try to test the limits where you encounter a good deal of unsettling unpredictability .

I've pulled supercharger data and found discrepancies between the before and after kwh delivered -- explaining some discrepancies in the charging and forecasts. I don't think anyone else has gone into that level of detail.

If the top of your charge level is the "last 20%", then you can have two or three or four or more "last 20%"s by buying a bigger battery.

I'd say the opposite, that you're losing the first 20% as your battery degrades, which is the least important part. And if you need a specific range for some trip, again just size your battery appropriately. Make sure 80% or 70% of the brand new range is enough for that trip.

Depending on the car brand you can drive several dozen kilometers even after hitting 0%.

Fully emptying an electric car is a boring exercise in stupidity. Unless you are going out of your way to do it, you will never run into a situation where you need to get your car towed.

That’s not true. I have documented a forecasting calibration issue specific to certain supercharging scenarios where the forecast drops 10% upon departure . The driver can follow the plan and still go below 0

"Electric owners under-value the last 20% of their battery. That is the most important 20%"

It's not exactly like an ICE. An ICE (in Europe) will put up a warning light at "I have roughly 50 miles left" you put your max speed at 50mph and find a garage. Job done.

I've owned quite a few of them over 30 years. I currently have a MG4 (Saic) Long Range and an elderly Renault Clio. The EV has a demonstrable range of at least 300 miles. UK temperatures. The Clio can do around 600 miles on a full tank.

I treat my EV in a similar way to my ICE. When it says it has 50 miles left, I look for petrol or sparks. That works here, now in the UK. It will work in quite a lot of Europe, some of the US, and will be laughable in most of Australia and Canada, most of Africa, ... anyway you get the idea.

They - ICE and EV are simply different. You have to learn to work with how they operate.

Oh and I pay £1.28 per litre and 7p per KWh (for overnight car charging). The petrol price is low at the moment for here (it hit £2+ when Russia went mental).

Quite loose numbers:

600 mi using 50 litres at £1.30 per litre is £65. 300 mi using 70 kWh at £0.07 per kWh is £4.90 say £5.

So, less than 20% cost in fuel (£65 vs £10) for an EV (here and now). I'm not too sure the Clio can really manage 600 miles nowadays but it is a good 15 years old!

If I have to use a motorway/commercial charger then the cost is around £0.50 to £0.80 per kWh (https://www.gridserve.com/electric-vehicle-charging/our-pric...) I don't use them very often

The EV will need a new battery in around four years time, or I pass it on.

ICEs are around 150 years old. I went to school in Abingdon, where the Morris Garage (MG) operated from in about 1930(?/ish). My MG is a Chinese effort and about as British as I am. EVs are at about the stage that ICEs were when a bloke had to walk in front with a red flag.

> Oh and I pay £1.28 per litre and 7p per KWh (for overnight car charging). The petrol price is low at the moment for here (it hit £2+ when Russia went mental).

I (in the US, California) pay less for petrol than you and more for electricity; 87p/L and 23p/kWh overnight.

That might be where the EV wheels fall off!

CA is notionally EV friendly but it is also a US state and it is rather large, so range is king. CA is mostly very warm so that helps with range. Gas (petrol) is relatively cheap and 'leccy is quite expensive.

The UK extracts rather a lot of tax (fuel duty) on top of the actual market cost of petrol. That's why our petrol/gas cost is pretty extreme. I have no idea how we ended up in this pickle (I have a few ideas), given that we have the North Sea oil deposits nearby. I think we failed to work that as well as Norway did. Bloody amateurs!

You probably have rather more land than me and could consider solar cells. My house conveniently faces south but its a two storey bungalow with three dormers, which means I can't put PV cells on the roof, facing the sun. My garden is also rather unsuitable for PV, being about 1/2 acre with mostly a 30% slope (its quite odd).

The world's climate woes are not yours or my responsibility. PV and EV may help or not. I can manage EV but not PV. You may find that the capital cost of deploying PV in CA might pay off quite quickly.

My IT company has a customer ... . They have a PV (solar panel) deployment on top of a building. This is in Dorset (UK). There are something like 20-30 panels on the roof. I've seen the monitor on a fairly bad day - 2 KW and in bright sunshine something like 8 KW. I'm pretty sure that CA could do rather better.

My total lot is under 5000 square feet (about 450 square meters), much of which is covered with house. I do have rooftop solar.

A few other notes:

- On average, both petrol and electricity is more expensive in CA than most of the US

- Petrol is expensive throughout the state, but electricity can be cheap if you live in a city with public utilities (The city of Sacramento has a public utility and rates are about half what they are on a private utility where I am).

- Peak rates are about $0.55/kWh (41p), but off-peak times are a generous 19 hours a day 21:00-16:00) in the summer, mainly because there is so much solar, so we have the "duck curve[1]" here.

- I live near the coast, so get fog in May and June, but July and August I generate quite a bit (over 10kWh per day this July).

1: https://en.wikipedia.org/wiki/Duck_curve

You note that you have rooftop solar - how is that going?

I'd love to deploy it but the south facing side of my house has three dormers on it which means: I'm stuffed. I can't fit panels on a 12m (40') long roof, with inclusions.

I wish you all the best, mate.

You're highlighting something that's not relevant. The people who are abandoning their EVs are doing so in spite of the tremendous savings. That should raise alarms.

Tremendous savings? I save less than 10% charging at home, at off-peak, vs a Prius for highway driving (3.8mi/kWh vs 50 miles/gallon):

  You have: (0.28USD/kWh)/(3.8miles/kWh)
  You want: USD/mile
   * 0.073684211
   / 13.571429
  You have: (4USD/gallon)/(50miles/gallon)
  You want: USD/mile
   * 0.08
   / 12.5

There are 3.8 litres to the US gallon. You pay $4.00 per gallon and I pay 3.8 * £1.30 = £4.94. I remember the days when it seemed like fuel was almost free in the US compared to here!

Your electric rate looks high at $0.28 per kWh. Its still less than I pay in general - £0.287 but I get an overnight rate at £0.07 (23:30 to 05:30).

It seems reasonable to have a low rate at night and always has been. In the UK it used to be known as "Economy 7" (seven hours not 7p!) That's nothing new and I remember it being a thing from the eighties. I'm surprised you don't have something similar.

I misremembered the cost for electricity; it's $0.26 not $0.28 (and that is the discounted overnight rate). Peak usage is over $0.50, but that's only 4pm-9pm in the summers.

The bill is rather maddening because there is no one place where they put the price per kWh. There is "Delivery Cost" "Generation Cost" and "Non bypassable charges" each with different values for peak/off-peak so I have to do the math myself.

Bloody numbers! I also have snags dealing with them and I have an A level in Maths.

Your discount rate at $0.26 - I assume that is for overnight usage - sounds a bit nasty. Are you able to move to another provider or are you tied to a single provider?

If the latter, then that is another discussion 8)

Basically one provider.

More details: All but the generation charges are from a single company that (among other things) is paying off a lawsuit for having not properly maintained lines running through highly flammable forests.

The generation charges can com from two companies; one has a slightly higher sticker-price, but discounts down to what the other charges, so the rates are identical.

I had my overnight electricity rate wrong it us $0.2608 which makes it about 6.8 cents per mile vs 8 cents.

agree

here on the other coast, with cheap gas and expensive electric, Prius would be about 6 cents/mi, Bolt costs me about 8 (on average winter/summer)

gas is $2.80/gallon, electric (with off peak program) $0.30/kWh. Bolt (fairly efficient EV) 3.8 mi/kWh (3.4- Winter 4+ summer)

Yup, in CA with PG&E a gas car is cheaper to drive than EV unless one has solar or free charging at work.

sure but here's an example from my mazda 3. after 17 years it exceeded it's rated range . the original full tank range was 337.5 miles. After 17 years i was getting 398.25 miles

With EVs you have a massive loss of utility over just a few years . Going from 337 miles to 320 miles in the first year is a huge loss. and down to 269mi after 10 years is catastrophic

> 269mi after 10 years is catastrophic

I have taken multiple 1000+ mi road trips with great ease on 280mi range. I would describe it as the exact opposite of catastrophic. And they've only become easier since then (e.g. more charger deployments).

When I bought my Prius in 2010, the longevity of the traction battery was a concern. A month ago, when I finally sold it to a garage, that car had over 190,000 miles on the odometer. I sold it because the transmission needed some repair work. The battery and the engine were still going strong.

To be fair, I’m pretty sure the 2010 Prius uses a completely different battery chemistry than modern electric cars. (I think it’s NiCad, but I’m not 100% sure on that.)

NiMH https://en.wikipedia.org/wiki/Toyota_Prius_(XW30)

NiCads suck, you find NiCads in like old AA rechargeables and cheap toothbrushes.

I don't know for sure if NiMH last longer than Li-Ion but I've had much the same experience with my Prius - Old as hell and everything but the battery failing

My prius (2006 model) finally had the traction battery (NiMh) start to loose modules at about 250K miles. It was clearly getting weaker, but drivable at that time. Then Covid hit, and it sat for 2 months without being driven / charge cycled. That pushed it over the edge.

That isn't predictive at all of NMC or LFP chemistries though (and I'm not going through multiple charge cycles per drive), but a fun anecdote. It was an entertaining project opening up the battery pack and identifying/replacing the bad modules.

In the end, other parts of the car were dying too, and the final straw was California's refusal to allow aftermarket catalytic converter replacements, and the Toyota's price (with no competition) was more than the vehicle was worth.

So far my two EVs, both NMC chemistry (Kia and Rivian) are at 80,000 and 30,000 miles respectively, with no noticeable degradation.

My Prius is also a 2006 but with only 194,000 kms. The first battery module failed 21 months ago. I bought a used one on eBay for $35 and it's been running perfectly fine ever since. I keep it undercoated but the body is going to go before anything else because we salt our roads in the winter.

Same for ours, a 2010 model. It's 15 years old and the battery and electronics work fine, even the GPS system. It's the other stuff that needs fixing: A/C, exhaust, various pumps, etc.

NiMH

I think batteries aging as much as they do is a thing we will be able to solve. Dr. Jeff Dahn's talks on YouTube were really striking in how much a bit of chemistry and charging management gets you.

https://youtu.be/i31x5JW361k?si=JdjJD_Lzg4qsY84C

His latest visit to the channel was 12 days ago:

https://youtu.be/oc3XgCfQfXs

>> That’s not bad, given that most cars are scrapped somewhere in the 150,000 to 200,000 miles range. At that point, a Tesla will have more than 80% of its initial capacity, and in some cases, even more. So people will probably give up their car, well, well before the battery gets close to becoming a burden.

Can they not see that this is because of correlation and not causation. Why would an EV be given up at 150 - 200K when it has much less moving parts and stressors compared to the traditional ICE based vehicles?

Some things will still add up. Someone who might be shopping in the price range for a used car with 100,000 miles might also see a car with 200,000 miles that needs brakes (probably for the first time in an EV's life), shocks, bushings, CV joints, A/C service, or possibly corrosion repair/mitigation in some climates, might choose to trade or scrap over spending those repair costs.

Also there becomes a crossover point of residual value where a car involved in an accident becomes cheaper to total than to repair, which is probably what takes a lot of cars off the road.

That mileage may stretch longer if the important parts of an average EV drivetrain can run without major service for significantly longer than the average ICE drivetrain, which seems like a likely possibility.

Rust, collision, part availability, and newer safety tech are all reasons to scrap an old EV. I hope manufacturers realize this and make the battery easy for DIY removal, similar to removing the catalytic converter from your rusty and bent ICE vehicle is the big moneymaker.

As a low mile driver with a tendency to hold on to cars, it would also probably surprise people how much the average life of plastics are closer to the 15-20 year mark, especially when regularly handled by people. Things like plastic knobs and buttons can break in interesting ways that a lot of people wouldn't expect. We tend to think of plastics as "forever chemicals" because we hear that term a lot, but it's not that plastic is particularly rugged against regular use across time, it's that how it breaks down is awful (see all the discussions of microplastics; it break down and then becomes a part of ecosystems in disturbing ways).

(ETA: Also the EV is so much more the "software-defined" car than anything, and the lifecycle of software versus tech debt and long term maintenance is going to be a large issue, even though the cars are mechanically simpler, the software is something making up for that in its complexity.)

> I hope manufacturers realize this and make the battery easy for DIY removal

This seems to be the case so far. A lot of scrapped cars' batteries seem to be going directly into second use in a second car. A lot of the manufacturers are also prepared for future "power wall" secondary uses of depleted batteries, but so far there has been too much of a market for the used batteries in second cars for used (even depleted) batteries to build a "power wall" market for used batteries. (Tesla's brand of that concept that sounds a lot like the generic term so far has almost exclusively been using new batteries for their products. Nissan's brand that no one has ever heard of, dedicated to used batteries only, has scarcely built or sold anything and is in danger of shutting down as an effort.)

The economics of used EV batteries is already a fascinating thing to watch, and something we'll probably see get more interesting rather than less.

Your remark about the plastics rings so true from my experiences. A friend's old Corolla used to constantly have plastic parts just crumble when handled in a regular fashion. New old stock parts would be nearly as brittle, often breaking only a few dozen uses. The car's motor was OK but everything else about the car was just falling to pieces.

Which is why those catalytic converters were stolen so often. I'm sure a battery would be a lot heavier and bulkier but I'm not looking forward to people stealing those....

When you’re trying to steal half the weight of the car that is likely integrated into the frame, it’s a hell of a lot more work than a quick 15 second cat grab.

And a lot more like by be lethal too.

Batteries are not like gas tanks. 80% of original capacity doesn't mean you have 20% less, it means a lot of other things too. These are chemical reactions which happen which lead to lots of other effects, like higher internal resistance.

Is this true!?

So does it mean that if I use a 80% capacity battery my actual functional value that I get out of it would be considerably less than what the 80% would infer?

"The degradation rate of lithium-ion battery is not a linear process with respect tonumber of cycles, battery aging tests (Fig. 1) have shown that in cycling tests the degradation rate is signiﬁcantly higher during the early cycles than during the later cycles, and then increases rapidly when reaching the end of life (EoL)."

Actually, 80% is considered effectively 'end of life':

"Battery end of life is typically deﬁned as the point at which the battery only provides 80% of its rated maximum capacity"

* https://www.researchgate.net/publication/303890624_Modeling_...

Funny because 150k miles is about the low end for the cars I shop for.

For the comparison to phone batteries, it's worth noting that until the last few years few if any phones allowed capping charge below 100% but now iPhones (15 and higher?), many or all Samsungs and unknown others allow capping for battery health. For the Samsung devices, the ability to do this came somewhere around the time they started promising 5 years of OS and/or security updates.

I'm pretty sure Tesla early on 'sold' an optional range extension that simply allowed you to charge the batteries further for extra range, with part of that cost presumably covering an anticipated higher battery failure rate. IIRC there were also some times when there were hurricanes coming during which they OTA unlocked that for everyone in the affected regions as well to facilitate evacuations.

> For the comparison to phone batteries, it's worth noting that until the last few years few if any phones allowed capping charge below 100% but now iPhones (15 and higher?), many or all Samsungs and unknown others allow capping for battery health.

Apple also added that the the M4 iPad Pro and M2 iPad Air. For those with an iPhone or iPad that does not support this there is a way to hack it. I got this from a comment here on HN many months ago, but don't remember who wrote it.

Apple Shortcuts automations can be trigged by battery level. Plug your charger into a smart outlet that can be controlled from Shortcuts, and simply make an automation that runs when the battery level goes above 80% that turns off the outlet.

If you want to even more closely match what Apple does on the devices with the 80% limit support built in also make a shortcut that turns on the outlet when the battery level goes below 75%.

Some smart plugs work better with Shortcuts than others. For example for an Amazon smart plug by understanding is that what you would have to do is make a routine to turn off the plug in the Alexa app. Then you would have to use a third party service like IFTTT to invoke the Alexa routine from a Shortcut.

A TP-Link Tapo plug is a little easier. You still have to set up the plug in the Tapo app, but the Tapo app includes Shortcuts integration so Shortcut actions to turn Tapo devices on and off automatically appear in Shortcuts.

That's still not perfect because you need to be logged into the Tapo app. I find that I occasionally get logged out (not sure if it just periodically logs you out or it happens on app updates) and then the Shortcut breaks until you log back in. I've found two ways to deal with that:

1. If it breaks the consequence is that night my iPad charges to 100%. You are supposed to do that occasionally anyway to keep the battery level indicator calibrated, and this is a convenient way to do that. If I wake up and the iPad is at 100%, I go log into the Tapo app.

2. I've got a Shortcut on my desktop to turn on the Tapo plug. I use that to turn it on before charging the iPad. If it doesn't turn on I know it is time to log in again. This way I only get the "fails to turn off" situation if I get logged out between the time I go to bed and the time the iPad passes 80%. I handle the need to occasionally charge to 100% by noticing when my iPhone 15 charges to 100% and then plugging the iPad into a non-smart outlet that night.

Best would be to get a plug that works with Shortcuts without needing to use the vendor's app and without needing an internet connection.

I thought that would be the Tapo because they can be set up as Matter devices, but after buying it I learned that this requires a Matter contoller and maybe a Thread border router. I have a couple of Amazon Echo devices that may be Matter controllers, but it turns out you need a controller for each platform (Alexa, Google, HomeKit) that you want to control the device from, so my Echos would not help with using a Matter device from Shortcuts.

My next step on the Matter path when I get some time is to look at Home Assistant. My understanding is that it can act as a Matter controller, and that the Home Assistant mobile app can has Shortcut integration on iOS. That should let me use the Tapo in Matter mode, and to replace the Tapo app and its annoying login requirement with the Home Assistant app.

There's also something called the "Chargie" (chargie.org) which is basically the same capability but connected on the USB side of things. I assume it's Bluetooth controlled, and it seems to have a bunch of other charge control options including rate limiting. Probably a lot more important back in 2019 when they released it, and probably relevant now for people who want more control (or laptop control with one model, though I suspect a lot of USB-PD laptops can cap charge and you might not want the laptop cycling power on and off).

Does not work well if you use multi-port chargers, i definite don't want my charger stop charging my laptop when my phone is full

Great post. Minor quibble: the data shows fast DC charging does not have a material impact on battery pack health longevity.

TLDR These batteries are going to outlast the vehicle chassis.

Full Speed Ahead: EV Study Reveals Impacts of Fast Charging - https://news.ycombinator.com/item?id=37330024 - August 2023

>This is a fairly common fear for people considering a new EV: “Won’t the battery need to be replaced after a few years?”. And I think it’s even more prominent in the second-hand market: “Oh, I’d never buy a second-hand battery!”.

I will admit that both of these are nagging on me. I fully intend for my next car to be an EV, but if I was buying today, this would be a factor. I drive a 2013 Camry (that I got used) that shows no signs of slowing down. I hope to drive it for at least a few more years. If the car is still reliable when it's time to send a kid in it to college, that's probably when I'll start looking for something new. And you can show me studies all day long, but my irrational brain is just worried that I won't be able to get 15+ years out of an EV because there just aren't that many 15 year old EVs driving around today.

Don't forget that internal combustions engines lose power and efficiency over their lifetimes. Bearings, piston rings and other components wear, injectors and valves get dirty, surfaces develop varnish, etc. My last ICE car started needing a quart of oil every few months and that was with very good maintenance and not being driven hard.

I've been curious about how the degradation compares to EVs. I'm aware it's different kind of wear and that there's different ways to mitigate and repair EVs vs ICE, but they both have their own lifetimes and loss of performance.

I believe the difference between ICE degradation and EV degradation is that the EV one actually affects the car's range.

While it is true that your car might consume more oil, and some other component might need replacing, its range, assuming it has been serviced properly, should be similar to what you could get out of it new.

I do wonder if the sum of the costs of getting the ICE car back to mint condition will be the same as getting some cells replaced so you get full range again.

> While it is true that your car might consume more oil, and some other component might need replacing, its range, assuming it has been serviced properly,

Well, until it dies completely (or to the point that servicing it would be more expensive to repair than replace). Then it's range abruptly drops to 0. We won't know for sure until we have more older EVs, but it may well be that EVs last much longer than that at 70-80% range. Which, especially if starting ranges increase, may be a very useful amount of range.

They definitely loose fuel efficiency over time, so they go less far on the same filled fuel tank. Its not as dramatic as a 20% loss but its not nothing either.

It's true we can't shake mainstream obsession with range, but I also think most people are a bit hesitant to take their 175,000 mile gasoline cars on long road trips. Not because of range, but because it just might break.

So old EVs can be just like old gas cars - used around town rather than for long road trips.

We road tripped our 2005 240K+ mile CR-V 750 miles each way every Christmas without a worry. We’d still be road-tripping in that if a negligent Subaru driver hadn’t rear-ended us and pushed us into a Prius ahead as the middle car in a sandwich.

The car before that was a 1998 Mercedes diesel with 225K+ miles on it that retired only because of body rust not mechanicals.

It helps that I did all the maintenance, so I knew how reliable they were.

Cars are insanely reliable and people get irrationally fearful when a car turns 100K and then again at 200K.

Agreed. I don't suspect "you are most people" but you can try to convince me otherwise.

What I said was what I think most people do. Not what is possible.

Do a lot of people sell a car that’s never once left them stranded at under 175K miles out of concern? Sure.

But I think most people who daily drive 175K mile cars would rely on them for a road trip without much consternation.

Almost no one does maintenance regular enough to have cars that haven’t grenaded something well before that point.

I know people that have literally never changed their oil.

One plus for electric cars though, fewer side effects for that kind of thing I guess?

I own a 2018 Model S with ~140k miles on it. I have primarily Supercharged it, and have driven it across the continental US several times. It has only lost 8-10% of original range. I get it, lifecycle anxiety is to be expected, but the evidence is fairly robust these batteries will last (and at least in the case of Tesla and my use case, I have an aftermarket person I work with in North Carolina who can provide me refurbished packs if needed).

Here is a 2018 Model S with 400k miles on it, although it's original battery was replaced under warranty: https://insideevs.com/news/717654/tesla-model-s-400k-mile-ba...

(I tried to import a BYD vehicle to the US, with an unfavorable outcome)

Your car is only 70% of the way through it's nominal lifespan. It seems like battery life is holding up well, but we'll find out a lot more as many of these cars begin their second decade of service, quite often with less rigorous maintenance. I suspect many/most EVs will make it to 300k and 12 years, but the oldest (truly) mass produced Model S are only just now turning 10 years old.

Certainly, we should keep collecting more data, but the longevity and lifecycle trajectory is obvious.

https://spectrum.ieee.org/ev-battery-life

https://www.nature.com/articles/s41560-024-01698-1

https://old.reddit.com/r/electricvehicles/comments/1jvwi14/g...

https://www.thejubjubbirds.com/hit-and-run-on-the-energy-tra...

> Just purchased a 2015 Tesla Model S 70D for $9k (USD). It was very worth it. It still holds about 88% of its charge after 175k miles. There are also some positive factors you didn’t mention.

The top commenter from the post just purchased a 10 year old EV that they judge to be perfectly good and unlikely to die on them soon.

I do think the anxiety about batteries is somewhat justified today, because the capacities are small enough that only have 80% capacity available could be a problem. But once the batteries are larger, I suspect EVs will actually last significantly longer than ICE cars on average.

My other quibble is when the author says the majority of cars are scrapped at 150k-200k..if this excludes wrecked vehicles, I suspect most are sold to used markets, even foreign used markets, not scrapped.

Not so much any more.[1] US used vehicle exports are down at least 2/3 since 2008. China is making so many cheap new cars that used US cars are no longer needed.

Most scrapped cars in the US are chopped up into little pieces, run through a separator for steel, aluminum, and everything else, and end up at a steel mill to be made into new steel. In Silicon Valley, the chopping and initial separation plant is at the Port of Redwood City.

[1] https://www.trade.gov/data-visualization/used-vehicle-trade-...

> But the types and structures of electric car and mobile phone batteries are not the same. Car batteries are designed to last far longer.

What prevents the same advancements from being applied to phone batteries?

Phones don’t have active cooling, or just proper cooling. Heavy usage and charging will heat up the battery a lot. A warm battery at a high state of charge will deteriorate fast.

Also, they aren’t designed to last very long. One manufacturer could reduce the max voltage in the BMS (battery management system), or reduce the charging speed at high states of charges, to significantly increase the lifetime of the battery but they somehow don’t. Though you start to see some phone limiting their charge.

Battery degradation isn't really a function of use as much as its a function of abuse.

Phone's are designed to drive their batteries hard. If you are gentle about charging it slowly and keeping it in the 30-80% state of charge region it will last a long time.

The worst thing you can do is aggressively charge it to full and drain it till dead constantly. It will last a few hundred cycles before it gets seriously degraded.

A lot of people unfortunately (fortunate for smartphone companies) do exactly that (you know that person in your life who's phone is always dead?).

Alternatively, a lot of people destroy their phone/tablet battery by leaving it plugged in all the time with 100% charge.

Phone batteries are cheaper so less motive to invest in changing the supply chains for a different design.

Planned Obsolescence.

I have a cheap 2nd gen Nissan Leaf. It's one of the worst cars for battery degradation since it has no thermal management. I'm in a temperate climate so that helps. After 6 years it's now at about 84% state of health. Not great but it's not catastrophically bad like some anti-EV people will have you believe. I believe the first year experiences the steepest drop in health. Maybe after another 6 years it'll get to 70% and it'll probably get sold to a person who drives shorter distances.

I don't know how it is in Europe, but in the US, a long warranty does not necessarily indicate a belief on the manufacturer that the product will last that long.

In many cases warranties are non-transferable (or only partially transferable), so a long warranty could just mean "We think you will sell this before it fails" rather than "We think this will last a long time."

This isn't a thing, and as a concrete counter-example:

https://www.tesla.com/support/vehicle-warranty

"Yes. Your New Vehicle Limited Warranty will follow your vehicle and be transferred to the new owner when a vehicle ownership transfer is performed through Tesla."

Does this imply that everyone who sells a Tesla privately voids the warranty?

No, when you privately sell your vehicle, you transfer it through Tesla.

I've never seen this on a car, which manufacturers have a non-transferable warranty?

Most important point is comparing it to loss of efficiency in gas cars. There's a lot more variance there, given the work that a gas engine done and all the ways it can be maintained (and lack thereof), but most numbers I've seen point to around 10-15% after 100k miles.

I track gas added to all of our cars (because my dad and his dad did). I’ve driven several of them to over 130k miles and one to 242k miles. I’ve never seen even a 5% degradation in mileage from wear. (I did see the ~3% drop in mileage when ethanol was added to the standard gasoline mix. I wonder if someone is confusing that for wear.)

If I had a 10% loss in fuel economy, I’d be looking for something wrong and fixing it.

There's a lot of this incorrect info floating around in the EV community. I recall one person, a year ago I think, trying to claim that gas cars won't start in the cold.

Not -50F either. 10F and such.

Similarly I've been told my EV won't last a day when the winter comes. I live in Houston lol.

I really don't consider myself part of any "EV community"; I'm an EV owner who has owned/bought several gas cars prior (most bought new, so I saw the efficiency drop over time). That said, most of my cars were of the highly efficient variety, Honda Civics and the like, so it may be more evident than in trucks etc.

The common retort I suppose is that if my efficiency went from 38mpg to 33mpg that I must be doing it wrong, similar to how a discussion on location of chargers devolves into a highway driving being an endurance sport, where everyone gets 600 miles of range, fueling up their cars in 3 minutes, with no need for any form of urination.

Gas cars do have trouble starting at cold temperatures though? Ironically it's often due to the starter battery.

Gas cars have absolutely zero issues starting in the cold, unless the battery is not good enough, or the car has another mechanical issue. EG, the car is broken.

To say it's because of the cold, is like saying they have trouble driving on flat tires too.

As a Canadian, who's been driving cars down to -45C mornings for decades, they just start. Older cars, and I don't mean "cars that have aged" but instead "cars from the 60s or 70s" sometimes needed a "block heater", which was basically just a fat resister you'd plug into AC. It kept the engine block warmer.

However I must reiterate... that's not commonly required any more. Not for 50 years.

Diesel cars don't run very well in the cold. They still work though unless it's super cold.

Diesel cars work perfectly in the cold. Once an engine has warmed up (half a minute), then it's the same for the car as if it was 30C outside.

I can second this, and do the same math and tracking (someday maybe cars will reliably do this themselves). The same can be done for electrics (power paid for and delivered to the car versus the miles driven).

When I was a kid ('70s/'80s) a car engine might die due to cylinder wear, burning oil and losing compression. I wonder if those might have been noticeably inefficient (say one cylinder of eight still ingesting fuel but not compressing fully and leaking exhaust products into the crankcase). Now I have an EV (fairly new) and an ICE car w/ 220k miles. The ICE car is leaking oil and needs some suspension work but I think it's efficiency is pretty much the same as it has always been.

I rebuilt the engine in my 1961 truck 3 years ago, the bores were worn enough it was noticeably down on power. I can't easily track MPG in it for a precise number (no working odometer), but the mileage increase was significant enough to notice a difference at the gas pump, I'd estimate a 4-5mpg improvement. This would be an extreme case though, I really don't know how that engine even still had enough compression to start. The ring end-gap was slightly over 1/8" (0.128"), spec is 0.016", so on the extreme end of engine wear.

To get back to EVs though, I'm not really sure they will last any longer than current ICE cars. Engine reliability has gotten good enough that a worn engine normally isn't the reason a car gets taken off the road. IME the main killer is either body rust or just too many small parts being worn out to where it isn't cost effective to keep repairing. Suspension parts will wear faster on an EV, since they're heavier than equivalent size ICE cars. I've driven a lot of mechanic specials over the years, and of the 7 cars I've sold to salvage yards only 2 were due to engine issues, the rest were either body rust making them unsafe or just too many things wearing out.

The kind of person that pays that level of attention is certainly on the far end of the bell curve, as is the person who guns their engine everywhere they go and goes 12,000 miles between oil changes and tire rotations.

> If I had a 10% loss in fuel economy, I’d be looking for something wrong and fixing it.

Have you never had a car start burning oil due to gasket failures and what not?

All of the cars I've owned had 80k plus miles on them when I bought them, and most recently I disposed of a 2001 Corolla with 265k miles on it. They all had various moving parts that failed at least annually and prevented me from using the vehicle *at all*. I'll take a slightly reduced range on my EV over that, any day.

I’ve had cars that burn/use oil. That doesn’t significantly change fuel economy though.

15% efficiency loss doesn't sound that major. My car's currently averaging 7.9 l/100km. It it goes up to 9.0 l/100 km, it means that I need to buy an additional 5.5 litres of petrol over 500 km driven, which is around 10€.

I won't happen. Gas cars don't lose efficiency like that, so don't concern yourself.

It's not really a comparable concern in gas cars, though. Range matters a lot more in EV because the charging stations are much rarer than gas stations.

If you lose 15% range in a gas car, ok, you have to get off one exit earlier to refuel. No big deal. But if you lose 15% range in a electric car, that is sometimes the difference between being able to make it to the next charging station (especially DC fast charging) station or being stranded by the side of the highway and needing a specialty tow.

> Range matters a lot more in EV because the charging stations are much rarer than gas stations.

The classic topic in every EV "debate." Gas car drivers can't imagine having the equivalent of a 7 gallon tank. EV drivers can't imagine having a tank that isn't full every morning when you wake up.

It's not about whether it's full, but whether you can successfully make it to a faraway destination in a reasonable amount of time (& not adding hours & days to the trip).

It's not really a "debate", just a consideration. We recently got an EV, and generally love it, but long trips are more difficult. We ended up taking the gas car because the EV would've added 5-6 hours to the trip, due to a lack of fast chargers along the routes we needed to take and the 200-ish mile range of our car.

For around town, it's wonderful though.

My current EV is comparable to my last gas car in terms of range - about 300 miles (Kia EV6) vs 270 miles (Toyota C-HR) (I'm talking the miles I saw in my use, not the spec sheet)

In Norway, gas stations are now a lot rarer than fast charging stations. That was a fast transition. Some remove pumps to add chargers instead, some just close down.

It's more a matter of your most common use. The average person drives around 40 miles a day. For those in that circumstance, who live in a home where they can charge in the garage, they may never touch a DC charger for months at a time.

Interestingly enough, the lack of fast chargers actually minimizes the issue of efficiency loss, rather than exacerbate it. I drove my EV from Texas to California and back in 2022. It wasn't super convenient, but doable, based on the spacing of fast chargers on the freeway. I had to charge each time I came by one, due to the spacing of them (being every 100 miles or so, I may not be able to skip one confidently). In that situation, a 15% loss in efficiency wouldn't make a difference either way.

Citation needed. EV chargers are actually pretty common, at least in the UK. Jump in any EV and the satnav will show you all the chargers nearby and they're basically everywhere. Sure perhaps 7kw and not 150kw+ but even grocery stores have many of those 150+ now, even if they don't have their own petrol station.

Yes, I'm sure it depends on the locale. Sorry for not specifying. In the US, particularly more rural areas, it's difficult because we have such vast distances. On a long trip, there aren't enough DC fast charger OR level 2 chargers, the ones that are there are often broken and/or in use, and it is quite the adventure to plot a longer trip.

I'm curious about battery replacement (1.5%) compared to the rate of engine replacement for ICE vehicles.

My 200k+ kilometers diesel Fiat Punto is as efficient as new.

there are electric cars made more than 115 years ago with the original nickle iron batteries that still work, less capacity than lead batteries, but essentialy idestructable and imortal. there is no reason to equate high power density and low weight with a short life span, and in.the case of capacitive electrical energy storage, we are only waiting for higher energy density, ss it is a true solid state battery, though as every substance that can exist will.hold a charge, the number of possible candidates numbers in the trillions, more, but whatever, and there is an absolute inevitability to creating batteries that will serve humanitys transportation needs. thousands and thousands of smart people work on that every day.

Having studied battery lifetimes in an engineering context for a significant amount of time I've regularly wondered how much of the slow battery degradation in these car battery packs is "cheating".

That is how much of the battery capacity is hidden by the battery management system when the car is new and then slowly doled out as the battery ages to make for the appearance of very slow degradation even though the individual raw cells would be wearing out quite a bit faster? If this were true what you would see is after this excess capacity was exhausted would be battery capacity falling off a cliff eventually, though this data seems to show a couple hundred thousand miles of consistent capacity with no cliff.

SSDs do a similar thing for capacity and wear with a sizable proportion of capacity reserved to replace bad blocks as the SSD ages.

Whenever I make this comment almost everyone responding is just guessing about how I'm wrong and new chemistries are so much better, etc.

Yeah as far as I can see all the companies that study EV batteries and provide degradation reports etc. all do so by using the data from the manufacturer. I would trust data about battery degradation a lot more if the data came from an independent data logger, logging voltage and current.

> I've regularly wondered how much of the slow battery degradation in these car battery packs is "cheating".

Using the word "cheating" has a very negative valence, but it's not exactly a secret that EV batteries are not designed to use their full "raw" capacity. The manufacturer is quite clear that you should avoid charging to more than 80% on a regular basis as it will degrade the battery faster. What matters is not that the batteries are capable of some theoretical "raw" capacity but that the advertised capacity is correct, just like with SSDs. It doesn't strike me as cheating that SSDs have more capacity than what is advertised on the (proverbial) box.

I don't know the right word, scare quotes were to accommodate for that. If not cheating then at least misleading or avoiding disclosing the actual mechanics and degradation of the battery. To the tune of it might be possible a new car would actually have 50% more than the range it allows you to use to make it seem like the batteries degrade much slower than they do.

>The manufacturer is quite clear that you should avoid charging to more than 80% on a regular basis as it will degrade the battery faster.

This is one of the things that doesn't add up. If the article says you can drive a tesla 200,000 miles and still have a mid-80s percent of total battery capacity left, why are car manufacturers being so clear about charging patterns to "save" the battery? With the std deviation bars in the graph showing a pretty small distribution, it would seem charging behavior doesn't matter (of course there will be people who don't follow the guidelines and if so there should be an expected much wider distribution)

The facts from studying the mechanics of raw cells of earlier lithium chemistries, the advice from the vehicle manufacturers, and the data in this article do not add up.

And also we have things like this, openly demonstrating much larger capacities

>Tesla extended the range of some Florida vehicles for drivers to escape Hurricane Irma

In this case a selection of Tesla vehicles were temporarily "upgraded" from 60 kWh to 75 kWh, that's 25%!

https://www.theverge.com/2017/9/10/16283330/tesla-hurricane-...

That’s because 60/70/75kWh cars all had the same battery pack, with lower end models software-locked, not because of any “cheating” that you alledge.

Ok the value judgement is up to you but selling a bigger battery pack as a small one will result in exactly the kind of artificial longevity I'm talking about.

I'm a bit confused as you're saying this article refutes your hypothesis, right?

I'll also offer up an example. The Polestar 2 (prior to 2024) has an advertised 78 kWh battery, but also clearly only 75 kWh available for use. That's about 96% right from the factory. So presumably it's doing what you're saying, but it's also not a secret. It's also a way to prevent regular 100% charges from happening, which have proven to accelerate degradation.

Their data fit on the extracted (supposedly real) data between 100k and 300k km suggests that you could drive around the planet 5 times while losing only a few percent of total battery capacity and I don't believe that raw cells behave that way regardless of recent improvements.

Tbh I'm worrying about everything in my car except for battery degradation.

Inverter dying, charging circuit dying, the motor shorting out, some of the DC isolators going bad, charging port crapping out. Even with the battery, my concern isn't with degradation, but that one of the modules decides call it quits and that's all she wrote.

Also a lot of EV engineering brings to mind the eternal Douglas Adams quote: 'The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at or repair.'

All in all, I find the claim that EVs will outlast gasoline vehicles a rather bold and extraordinary claim, and one that won't stand the test of time.

> Inverter dying, charging circuit dying, the motor shorting out, some of the DC isolators going bad, charging port crapping out.

How many of those things are very settled technology though? ie. a lot of years spent iterating and manufacturing out points of failure. The battery itself is the newest technology, such is my understanding (feel free to correct me).

They feel like unnecessary worries to me. But then, unnecessary worries seem to be a defining trait of the nowadays.

> How many of those things are very settled technology though?

None? We simply don't have the kind of reliability data on EVs going out for 10-20 years that we reasonably expect a well-made ICE car to run for (with regular maintenance and without major component failures).

These failures are not theoretical, and have been demonstrated in the wild, and can easily total your vehicle. They might affect you personally if you plan on keeping the vehicle for the long time, or the resale value if you don't.

Inverters are known to degrade and fail, solar inverters basically almost always die somewhere between the 10-20 year mark (and those switch far less wattage and don't operate under conditions as harsh), which imo is not acceptable for a car to get totaled in that amount of time.

Unfortunately EVs are one of those partisan topics on the internet which have a lot of partisan haters and lovers, which makes reliable info on them almost impossible to find.

I personally have both solar at home and drive an EV, so I guess I both into the thing, but I can't say with high confidence that my EV will make it to the 20 year mark with the equal probability a well-made ICE car would.

My point was more that "electric systems" are a well entrenched technology / solved problem. As in, it's not a new method of power delivery. Though, admittedly, I'm not sure how much of that transfers directly into the context of powering a car.

As more time passes I guess we'll know which parts are more likely to fail and they'll be designed around the ability to replace them or design them for robustly, or both.

> I can't say with high confidence that my EV will make it to the 20 year mark with the equal probability a well-made ICE car would.

Fair enough.

I think it's quite positive that this comparison can start to be made, however. Given the relative immaturity of EVs.

EVs are quite unlike most residential power systems you encounter. They use high voltage DC, which this video illustrates, is a fun topic in of itself:

https://www.youtube.com/watch?v=Zez2r1RPpWY

To generate this, the batteries have more than a hundred cells wired in series, if any one them fails, that means your battery's dead.

In addition to this, the inverter that generates the AC driving your motor needs use high powered IGBT to switch this huge current and power tens of thousands of times per second to do PWM and uses filter caps to smooth the output. Both these are wear items(and impossible to replace) , its the operating conditions that determine if they'll live for 50 years or less than 10.

Which is why I think EV reliability will be no different from ICE - manufacturers who do give a damn will make them last long, and make them fixable once they break, while others will save a buck and won't.

The study data showing average capacity is helpful, but the lower quartile and even more so the bottom 10% is really what people worry about. In the used car market the presence of even a decidedly small number of “lemons” has a significantly detrimental price impact.

How can you tell how much battery capacity is left?

Any decent battery system measures the current that goes into the battery, and the current that goes out. Off-the-shelf ICs "learn" the battery's initial capacity and its state-of-charge to voltage curve, and thereon can observe degredation below those initial measurements, as well as fairly accurately reporting how much energy is in the battery at any given moment.

Huh? Battery has never forced me to upgrade. It has invariably been the inability to still use the "new and improved" version of google maps.

I use my phone instead. Having these apps in the car is like buying a smart TV and having the Netflix and YouTube apps fail after a few years due to changes in the services that the apps can't handle and the manufacturer not providing updates any longer.

Can we talk about battery safety?

Old lithium batteries are ticking time bombs. Swelling, leaking, and ready to ignite at the slightest spark or contact with moisture.

As their chemicals break down, one short-circuit can unleash a chain reaction of fire and toxic smoke. Even sitting forgotten in a drawer, they can suddenly swell, rupture, or explode without warning.

Reply on news.ycombinator.com