EV Batteries Are Defying Expectations After Miles | Hacker News Reader

EV Batteries Are Defying Expectations After Miles

wsj.com

76 pointsapparent93 comments

No manufacturer is testing the batteries life by just charging and discharging them daily for a decade before releasing them. Instead they are using artificial acceleration techniques like getting the battery hot while charging/discharging continuously to simulate a longer lifetime. They can't realistically do anything else to estimate it. But it turns out heat is the big enemy for li-ion batteries and if you can keep them on the cooler side of their range they will last a lot longer.

we have EVs for almost 15 years now, one would imagine some real world data on battery life was collected already, even if not widely shared

Battery chemistries have changed and technology has improved a lot in that time frame. Those 15 year old Nissan Leafs have terrible battery degradation, because they're not liquid cooled (hot = bad) and used batteries that degrade fast (LMO, which autocorrect thinks should be spelled LMAO). Manufacturers learned their lesson and every EV built in the last 5 years is far better on both counts.

The only thing you can fault them for is for how long it took to switch to liquid cooled.

The leafs are fine cars but you're paying for being an early adopter. I bought one used in 2016, used it as a daily commuter till 2025 and then sold it. I don't think any car will ever be as cheap to run as that one....

There was and publicly available, although the mainstream press mostly chose to ignore it.

My VW hybrid is just over 10 years. The advertising claimed 28 miles on a charge, but the reality was about 18 maximum, with stop-start somewhat less. After a couple of months when it had calibrated, it settled displaying 21 miles for a full charge. 10 years on, and it still reads 21 miles on a full charge, but the reading is less useful in use. It's common for it to drop from 21 to 15 miles after a mile, but to stay on 1 mile remaining for around 4 miles travelled. But essentially, I don't consider the range to have deteriorated too significantly, it still feels around the same ballpark.

A lot of the original thinking about batteries comes from the Nissan Leaf, introduced in 2009, when they guessed that batteries would probably need replacing after 10 years. However, from the cars that were written off from accidents, they discovered much less battery wear than predicted, and around the time I bought my car, they were suggesting batteries would be around 80% after 10 years and most manufacturers started using that figure. I'm not sure what current estimates are, but I wouldn't be surprised if it was closer to 90% after 10 years.

It's common knowledge amongst owners that range is significantly reduced in cold weather - I'd guess from my use, maybe 20% less range on a cold winter day compared to summer.

And as to why car batteries last so much longer than e.g. phone batteries, I think it's mostly because their charging limits are much more pessimistic. Typically a car charger won't ever charge beyond 90% capacity or discharge below 10%, and this is built into its estimated range (so 0 miles remaining is 10% charge).

I have enabled "charge up to maximum 80%" setting on my phone from day 1, and 2 years later the battery lasts noticeably less. maybe it would have been even worse without that setting.

I wonder if it's because phone thermals are way worse so always heating up?

YMMV. Likewise, I've been using charge to 80% from the start, the phone is approaching 5 years, and I can't say I'd notice the battery being any worse from new.

Bigger issue is that it's been out of OEM updates for 3 years now, which is complete balls considering it's a reputable brand and came with "Android Enterprise Recommended" or whatever that was.

“They can't realistically do anything else to estimate it …”

Of course they can.

How do you think they get the MTBF for hard drives?

What they do is deploy thousands of them and wait for the first one to fail. Then they wait for the second and third ones to fail… And then they can construct a statistical abstraction for the entire population based on the very first degradations … and that doesn’t take long at all.

If they want to avoid confounding variables to real world use, this could take years. Nobody wants to wait years after an innovation to sell their product, so they develop ways to speed up the testing and get a number.

A machine that presses a keyboard switch 500 times a second for several days straight is obviously not indicative of someone actually using the keyboard. But it'll get the "absolutely beat the snot out of it" number, which is usually good enough for marketing.

No, it doesn't take years and the testing is performed, roughly, as I just described it.

They build a test rig with 10,000 drives which they run continuously producing a total "drive hours" count and then wait for the first one or two or ten drives to fail.

X failures / 10,000,000 drive-hours will give you your MTBF, etc. ... and only takes ~40 days given a 10k drive test rig.

Some pictures from Seagate lab in Longmont:

https://cdn.mos.cms.futurecdn.net/yHqDagRDz9H5koWBefi24K-120...

https://cdn.mos.cms.futurecdn.net/EiZ9tA6p3UbaJFTaW75JdN.jpg

Yes, this is one of the hacks they use that are fundamentally inaccurate due to confounding variables like bathtub curves and different issues causing early failures and long-term failures...

Because, otherwise, they would need to run those drives for years and years with reasonable cycling and read/write pressure to report accurate numbers based on real world failure rates in real conditions.

Wasn’t the same concern with the Toyota Prius when it was first released? Only for all the doubters to be proven wrong by the taxi drivers who kept beating Priuses for decades.

In any case battery failure seems rare but it still is catastrophic and nobody can afford replacement. Hence companies should just provide some sort of warranty / insurance product for the few unlucky folks. Seems like an ideal candidate.

They do.

Generally enforced by regulation, particularly tied to any government grant program to encourage EV adoption.

Those old taxi Priuses have had their batteries replaced.

They were also using NiMH batteries - a completely different chemistry from modern lithium batteries.

Lithium EV batteries are regularly serviced and repaired?

The original Prius batteries can easily have individual cells replaced (mine is 20 years old, we've replaced 2 cells) - of course Toyota want to sell you an entire battery, the wrecker will sell you a whole bunch of spares

(do this with care, an entire pack is hundreds of volts)

Generally you can't replace cells unless you find perfect match - needs to be same resistance and capacity. Only on very large scale this is possible.

That said on most EVs blown cells are very very very rare. Mostly it's balancing lead that looses contact.

These are NiMH cells (7V) wired in series (not parallel so matching is less of an issue) - one bad cell and the series resistance of the whole string is shot

kia ora. interesting to know. so they are pretty small packs and their design doesn't really apply to large packs.

I replaced some of the NiMH cells in my Prius traction battery by buying some Prius traction battery cells on eBay for $20 each. Still working fine after 2 1/2 years and a lot cheaper than buying a refurbished traction battery for $3,000. My Prius is 2006 so any cells from the Gen II models (2004-2009) work fine.

Isn't this old news? I remember reading about 7yo teslas used exclusively in cold climates (Norway, Finland, etc) and they found the same thing: batteries held on much better than even the manufacturer expected. And those were often 1st gen cars, which you could expect to have teething issues.

It was at the time one of the main reasons the 2nd hand markets in those countries were pretty healthy and saw a lot of movement of used cars.

Some of the early models - either the Chevvy or Nissan Leaf - had battery packs which would degrade very quickly. This was causes by the charging system which seemed designed to destroy them.

The expectation in the EV fan world was that modern (> 2019) battery packs were very likely to retain 80% of their charge for well over the time anyone expected with the drop to 80% itself taking most of the car's expected lifetime. This was because the standard extrapolations ignored improvements in charging algorithms and that real-world usages were expected to be better for the battery than the models.

Another prediction we made is that range under cold (or very hot) weather would be significantly reduced and that seems to be the case too.

EV batteries may be lasting far longer than most buyers expected but are consumers really going to trust the data over the headlines? There has been such a propaganda attack on EVs that it will take a long time to overcome that.

In late 2022, I moved into a new semi-detached house in a small project of 33 such houses total. Back then, there was precisely 1 EV here, parked openly.

This has grown to 6 in the meantime. It seems that the segment was growing even before Hormuz, propaganda or no.

If batteries are actually lasting, then it's easy to counter the propaganda. Manufacturers simply need to make sure batteries are replaceable and offer guarantees that they'll be replaced on the manufacturer's dime if they degrade.

That's easy to say when you aren't the one considering adding a huge liability to the companies bottom line.

There are also some elements of EV ownership where they require some TLC, like not overcharging them when doing mostly local journeys, not using rapid charging more than necessary etc. If manufacturers were carrying the can it would be easy to let these things go out of the window.

Saying, "Oh, you just have to baby it and it'll probably be fine!" is probably not the best way to counter claims that EV batteries are unreliable.

Also, consumers currently are the ones taking on liability in adopting new technologies. That's one reason why EV adoption has been a gradual thing. Offer certainty to consumers and they'll look at what they have to gain instead of focusing on what they might lose.

> That's easy to say when you aren't the one considering adding a huge liability to the companies bottom line.

Obliging companies to make repairable products? Removing their right to lock everyone into disposability and overconsumption? And even worse - trimming a sliver of their profits? Damn right, buddy: we can't be having that.

Those are completely orthogonal to accepting all of the risk, from the point of view of a CEO, of battery failures for cars you sold 10 years ago.

But yes, making things repairable is really important. And, I certainly have concerns that the danger of high voltage batteries will soon be used to introduce so much regulatory burden that it becomes impossible for individuals and small companies to repair them.

The black PR has been very effective. I own an EV and the first question I always get from non-EV people when they find out is "…but what about the range, isn't that a problem?", while the second one is invariably "but what about the battery, I heard it only lasts a year or two and it's so expensive to replace!".

The myths are now strong and it will indeed take a long time to dispel them.

the propaganda works because we are all familiar with the smartphone battery degradation after 2 years, so it seems logical that same thing will happen with car batteries

> we are all familiar with the smartphone battery degradation after 2 years

I haven’t experienced this for years.

It was very common in the 2012-2018 era. Batteries have gotten larger (less cycles) and better (sustain more cycles before losing capacity) since then.

Also phones don't have the same sophisticated circuitry or devices as a car to keep batteries in an optimal state, e.g. Nissan Leaf gen 1, which had significant range losses (on top of having a small range to begin with).

This is very true. There's also a habit where people are reluctant to get their vehicles batteries low as well. Folks seem reluctant to get below 20%. I think it comes from learned behavior from smartphones.

I will happily arrive at a supercharger with 1%.

I mean, we have an EV and you do have to think about range if you're going over 100mi away and time is tight (eg you have young children, appointments, etc)

When I can throw 100mi into my car in 5 minutes without having to do backflips to get the charger to start, or drive out of my way to specific chargers, and not pay 20 times the market electricity rate... Then people will stop worrying about it.

I do think we'll get there with new EVs (800V platforms look amazing), but it's not going to be retroactive. There are lots of EVs in used circulation where range will be a conscious consideration for another decade or two.

The early generation Nissan Leafs (no battery cooling) did most of the work to tar EVs longevity. It's not propaganda. They were shit.

I am confused by your comment. Any EV with a 250+ mile range and your scenario is moot. You just go.

Had an EV since 2018, range is never a concern.

My only concern is that traditional American car manufacturers still suck at making EVs.

> Any EV...

Well if you're going to limit the universe to a set of cars that can drive far enough. Sure, new premium EVs bring far less range anxiety.

Used, budget EVs have both longevity and range issues. Used, budget ICE vehicles do not. It's important to remember that the market isn't just $50k+ vehicles.

And cars on the limit cause the most hand-wringing. Did you plug in last night? Do we stop early or at a meal time? You plan your day around it. It's not insurmountable but it exists.

We had much of the same skepticism and anti-EV propaganda here in Norway a decade ago, when EV's started becoming popular. The usual claims was that the batteries would have to be replaced every five years, the cars would stop working in the winters, the batteries are fire hazards, the power grid won't support everyone buying EV's, etc.. All of which turned out to not be true.

Today the market share for EV's is 98% and they account for the majority of the total number of cars on the road! The people who bough 1st. generation EV's as a number two car a decade ago chose EV's as their main car before long. My own vehicles are 7 and 12 years old and holding up well, despite the 12 year old having inferior battery chemistry compared to modern cars.

Change isn't always easy or smooth, but this one is inevitable.

98% of NEW EVs. But the total market share of the fleet is about 1/3. Impressive. If EV share is 90% in a decade I wonder if Norway will see fuel stations being torn down.

I'd imagine trucks would still be diesel, so it'd probably make sense just to remove some of the pumps and replace them with rapid chargers. I doubt the fuel stations themselves will disappear, just cater to a different market.

OTOH in the UK at least, I think public charging is way too expensive compared to home charging, so I think most public charging is done out of necessity on long journeys, so the number of people using chargers would probably be significantly lower than number of people filling up with fuel.

Electric semi trucks have appeared from all major European manufacturers in the last years and I expect them to become dominant even quicker than electric cars on simple cost grounds. Electricity is cheaper than diesel and due to favorable regulations in Europe, for many use cases there is little to no time lost charging and usually also no loss in load capacity.

Good point actually. I guess the maximum 8 driving hours per day works well if they can find a rest point near a charging station.

According to YouTuber Bjorn Nyland, some such as Circle-K had already installed charging stations and are now removing some pumps and replacing them with chargers while adding even more charging stalls.

Thanks. The video was interesting.

Yeah, it's amazing how fast the transition has been. Many gas stations are or have already reallocated space from gas pumps to charging stations, and new stations are designed to accommodate EV's. Take a look at this station as an example - it's already five years old at this point: https://tinyurl.com/4y8u36m2

I should have been more accurate in my "majority" claim; it refers to EV's being the most common type of car compared to gas or diesel. Source: https://tinyurl.com/ycke45dy

I'll just caution that new things are often over-built to some degree. Then "engineering" is often how much margin they can squeeze-out while still hitting their warrantee targets.

General Motors built an ignition switch for decades which worked mostly just fine. Then they built one so crappy that they killed a bunch of people. Only to save 25 cents or whatever.

EVs depend heavily on 'early adopter evangelism'. I wouldn't read too much into this.

Although I have had rechargeable batteries in gadgets for decades, I would not like to have an EV parked in a garage beneath my bedroom, juicing up with many kilowatts, every night.

No amount of fire exits would help, I mean, what about my stuff?!?

No amount of insurance would cover personal belongings that could get lost, so that means stress from the stupid EV.

The technology might be generally proven (the EV could be a Toyota Prius variant) but the current iteration could be another Note 7, there is no knowing for sure.

I assure you it is not propaganda or TikTok that makes me wary of EV batteries, or any grudge, it is just that I have a different set of circumstances to the guy with a separate garage for housing an EV a safe distance away.

It took an extra decade for hospitals to allow mobile phones on their premises, they just had a different risk assessment. It wasn't propaganda that made them enforce such a ban.

Hence, although it is tempting to blame 'enemy propaganda', there is a whole spectrum of risk assessment, based on different people and different circumstances, so it will take a while, with battery chemistry essentially static for a decade before some get won over.

I hope you do not have an ice vehicle parked in your house currently, as these tend to burn much more often…

You are aware that there are multiple different chemistries, right? Go look at people drilling into LFP batteries on YouTube.

ICE cars are not immune to fires either - there are lots of recalls due to faults in the 12V systems causing fires even when parked.

No one throws Duracell during riots for a reason. Where as a small bottle, half filled with petrol, and a lit rag makes a pretty good anti-tank weapon.

A car with a tank full of gas is basically a big bomb.

I'd rather sleep over the electric car.

I can't remember an automaker ever doing a recall for an ICE vehicle that would spontaneously combust while parked. Maybe it happened at some point (ICE vehicles have been around for a long time), but there have been several for EVs. The probably is undoubtedly caused in part by the fact that EVs are charged while parked, whereas ICE vehicles are generally just sitting there, turned off.

You can't remember it because it's so common it doesn't raise any media interest:

ICE vehicles: ~174,000–200,000 fires/year (1,500–1,530 per 100,000 sold).

EVs: ~25 fires per 100,000 sold (hundreds per year total).

Is this referring to recalls related to defective vehicles? Also, EVs are on average much newer than ICE vehicles. I would assume that most ICE vehicles that spontaneously combust are not newer ones.

> EV batteries aren’t without their challenges, though. A battery that is frequently fast-charged with high power loses its range, on average, at twice the rate of a battery charged at a lower power, according to telematics company Geotab.

I wish my car could charge at different rates depending on whether I'm going to need to use it again soon. If not, it could trickle charge and be full enough by the next morning.

Does you EV not have a charge rate setting for AC charging?

Or do you want such a feature when DC charging?

I charge over AC at home but just wish that I could specify the rate by telling the car/charger when I'm next going to need the car.

> When early EVs hit the market, buyers’ concerns were well-founded. Roughly one in 12 EVs built from 2011 to 2016 have had to have battery replacements. But new data shows that more modern EVs are doing better so far. Among EVs built from 2022 on, 0.3% have had battery replacements, according to a 2025 study from Recurrent.

This is kind of a dumb comparison. It would be more useful to know how many vehicles had to have replacements within X years of being put in service. Obviously newer vehicles are much less likely to have needed a replacement at this point in time.

They still die by calendar age based degradation. High miles low years isn't interesting. We know that works well. They don't like talking about the calendar age degradation. Every article like this leaves that part out. It's annoying. Many articles have been written like this. Many more will be written yet. I guess there are still people out there who don't know that EVs are ideal for drivers who accumulate high miles per year. Personally I don't think batteries are going to get interesting until solid state batteries. The problem is the electrolyte.

> They still die by calendar age based degradation

Source?

Not OP, but this talks about it: https://www.sciencedirect.com/science/article/pii/S2352152X2...

That article says calendar ageing was the dominant ageing mechanism for batteries in their test vehicle, because said vehicle spent 96% of its life stationary. Unless I'm missing it, the article doesn't put a number figure on the rate of calendar ageing.

Real-world observations suggest batteries are likely to be serviceable for around 20 years, which is around the same lifetime of an average ICE car. Users who can tolerate a much reduced range (which is most of us) can likely extend this even further.

It's not clear how much of a factor calendar age degradation is in EVs because almost nobody is straightforward about it like they are in the battery industry itself.

Lithium batteries age even just sitting on the shelf. I fly RC planes and we store our batteries at 3.8V to lengthen their life, but they still deteriorate even when not used. Like anything else, I guess.

The article specifically talks about how this has changed with the evolution of chemistry in Li based car batteries.

I suspect the RC plane batteries you've been using for five years are not the same chemically as the EV car batteries in use in the UK for five years.

I can't read the article, as it's paywalled, but they're all LiPo/Li-Ion batteries, what's the difference? Hell, most cars just use the same 18650 cells we use, just a lot more of them.

> but they're all LiPo/Li-Ion batteries, what's the difference?

As I mentioned, the article specifically mentions that the battery chemistry has been changed and improved.

It neglects to dive into the chemical engineering specifics of such changes, as would be expected for press such as the WSJ.

I dare say there is as much variation in the attributes of the LiPo/Li-Ion battery family as there is within the borosilicate glass or soda–lime glass families (ie heaps).

Sure, but why are you assuming RC planes don't use the chemistry improvements as they come out?

It’s a niche, small-volume market. I would be shocked if novel industrial battery chemistries are reaching it within even years of the frontier. (Less surprising would be small-batch novel chemistries. But these aren’t really comparable to the industrially-produced thing.)

We mostly use the same 18650 cells everyone else uses for everything, it's not like we're waiting for custom batteries.

Again, the article specifically talks about improvements in car battery life since changes were made to tweak the chemistry (very likely in small dope ways).

The fact that the article states that this appears to result in longer life and the fact that you talk about your RC plane batteries not having a longer life would suggest that the fine specifics of the two batteries may well differ.

IDK I rarely see 18650 cells in RC. It's mostly pouches which I assume come from smartphone/laptop industry - these are likely to have highest margins and highest tech (other than space).

Put them in the vegetable bin in the fridge, too. Should make them last much longer.

Can confirm, I bought some Panasonic cells roughly 10 years ago for a battery pack, and have ~12 cells that didn't make it into the pack.

They've been sitting unused, in their original packaging, never opened... They're still sitting at the charge they shipped at, but the capacity is so diminished that one can't even run an esp32 for a day. I've tried cycling them to see if I can get the capacity back up, but I think they're toast already.

Yeah, and those come at 3.8V (storage voltage) already, so it's not like they were sitting fully charged.

But these are not sitting on the shelf are they, they are in a car with all sorts of systems managing the battery (well Teslas do).

False

"A 2015 Model S with over 265,000 miles on the original battery (85% capacity remaining)"

That's 25000 miles per year, which is high. So the opposite of false.

The comment above is (I would guess) about the eleven years of original battery usage rather than the mileage.

The implication being that runs counter to the claim of "calendar age degradation".

Do you mean 11 years is a "long" time? I'd say it doesn't matter how long it was - it ended up better the expected based on the miles, and since it was a high-mileage-per-year car, that supports the claim that degradation is determined (at least more than expected) by calendar years, as cgyv.. claimed. I don't see how it doesn't support his claim.

It's all moot speculation in the absence of any firm notation of what cgyvbunji considers "calendar age based degradation" to be.

The battery industry defines calendar age degradation, not a random guy on the Internet like me. It's like saying you can't discuss voltage without me defining what voltage is.

So according to you, the more you degrade the battery, the longer it lasts?

Usually you would think that 25000 miles per year over 11 years would degrade the battery faster than the car just sitting around for 11 years.

According to cygv... the more intensely you use the battery, the more miles you get out of it, so kind of yea, the faster you use it, the "longer" it lasts in miles. And this example supports that idea because it was used more intensely than usual and lasted more miles than usual.

Yes, exactly, as long as you don't fast charge. The ideal scenario for TCO measured as dollars per mile driven should be driving every day until you are down to some threshold, slow charging overnight, and repeating that each day until it's a brick. Should end up with something like 200-300k miles that way.

OP said 'They still die by calendar age based degradation', they don't was my point. They degrade based on charge cycles.

My EV is also over ten years old and has no noticeable battery degradation. I'm sure there is some degradation that could be measured, but in daily driving, I have the same range I did the day I bought the car.

Do you have more information regarding age based degradation? I haven't looked too deeply into the topic and I am not sure if the argument had been about pure age based degradation or about "after X years because a person will have driven Y kilometers since then".

This is just untrue.

It is not necessarily untrue (source: am currently involved in battery R&D). However, the degree to which it is true will vary strongly based on the particular choice of chemistry, what sort of additives/agents are used (e.g. scavengers/getters), cell/pack engineering choices (protection against ingress of outside atmosphere and such), and so on and so forth.

If this is true it's interesting, as ICE cars typically prefer to be driven continuously at operating temperatures, rather than cold. This is why a high-mileage low-year car can actually have an engine in much better condition than low-mileage high year cars.

Reply on news.ycombinator.com