NHacker Next
  • new
  • past
  • show
  • ask
  • show
  • jobs
  • submit
Modern EV Batteries Rarely Fail: Study (insideevs.com)
PaulKeeble 9 days ago [-]
The projection on most of these batteries life times now is 500k+ miles before they hit 80% of their capacity. The car around the battery will be falling to pieces before they reach that point. Its not like there weren't quite a lot of ICE lemon vehicles with engine and other problems that caused issues, its always been part of any industry and there will still be packs in the future that don't perform as they should for as long as they are meant to.

EVs generally will require less maintenance as they are substantially simpler vehicles with less moving parts.

satiric 9 days ago [-]
Cells often get sold with a "# of charge/discharge cycles until you hit 80% capacity" rating, at least on the 18650 datasheets I've looked at. So my guess is that they took that number and multiplied by the range per cycle.

The problem is that (1) this is simplistic, and doesn't take into account temperature fluctuations, ac usage, etc, and (2) battery datasheets are essentially sales brochures, and often stretch the truth. (For instance, it's rare to see a battery that can actually hit the rated capacity under realistic conditions).

theshrike79 8 days ago [-]
And not all 18650 cells are equal, there's no way to know what's actually used.
dsalfdslfdsa 9 days ago [-]
That really depends on your definition of complexity. After all, EV charging points have far fewer moving parts than a gas pump, yet we know how reliable they are.
bryanlarsen 9 days ago [-]
I used to work at a gas station. They are not more reliable, not even close. The difference is that a gas pump is never unattended and the attendants can quickly do the easy fixes, like a reboot.

Pay attention to how often a single pump is out of service at a gas station. It's quite frequent. You just don't care because there are other pumps you can use.

OTOH, most non Tesla charging stations are a single charger with multiple heads. Redundancy is a big reason for the perceived reliability of gas stations and Tesla superchargers.

heeen2 9 days ago [-]
Gas pumps are operated by customers in many parts of the world
bryanlarsen 9 days ago [-]
But there's always a human around somewhere.
interstice 9 days ago [-]
There are quite a few unattended stations in NZ
PlattypusRex 9 days ago [-]
The problem isn't just the number of moving parts; it's far simpler to design a common gas pump than it is to design a DC fast charger, the electronics inside, the software, the network connectivity, app development, and, most importantly, the ability to successfully handshake with every single EV out there (easier said than done).
dsalfdslfdsa 8 days ago [-]
Exactly. It's possible to build an incredibly reliable mechanical system, and all-too-possible to build a very unreliable system with no moving parts and lots of software (and network connectivity).
LUmBULtERA 9 days ago [-]
The Tesla superchargers have had perfect reliability in my experience, and it seems others agree that they are highly reliable.
dsalfdslfdsa 9 days ago [-]
Right, so reliability is mostly unrelated to the percieved mechanical or physical complexity.
hacker_88 9 days ago [-]
Lack of vibration from Engine might also fare better for parts around the battery
SkyPuncher 9 days ago [-]
Engines vibrations aren’t really a problem on modern vehicles. You get some, but they’re incredibly tiny compared to conditions. Heck, even on my 25 year old car I don’t really have engine vibrations.

When I do get vibrations, it’s almost always from unbalanced wheels/tires.

robdar 8 days ago [-]
>EVs generally will require less maintenance as they are substantially simpler vehicles with less moving parts.

I keep hearing this, but then keep seeing stories about huge repair bills and wait times for repairs on EVs, and threats of insurance rates for EVs to increase because of the high cost to repair.

A large percentage of the repair work on fossil fuel cars is body work, brakes, suspension, etc. All components that EVs have. I’m skeptical that this work will be cheaper on EVs than any other car.

theshrike79 8 days ago [-]
The huge repair bills are in big part due to the fact that there are no spare parts.

Every part that can be manufactured is put into a new car and sold, there is very little surplus of parts. And what little is sold, is sold at a premium. Add a nonexistent 3rd party market to that and you've got a supply issue.

For my ICE car I could just go buy a generic part if I didn't care about the quality - and in some cases a cheaper part made for another brand (Skoda) that was 1:1 with the more expensive part (VW branded).

What doesn't break is the electric engine, there is one moving part, the shaft and excluding some outliers the engine will outlast the body of the vehicle many times over. Same with batteries, barring manufacturing issues or damage, the wear is very predictable and gradual.

This is what gives the (illusion of) reliability to EVs. There are very few mechanical parts that can just suddenly break and leave you stranded. Software and electronics though... =)

SV_BubbleTime 9 days ago [-]
I’m an automotive EE.

This may be a rating you heard somewhere but the truth with practical real-world use is… debatable.

If you run these in a lab, ok, maybe, according to someone trying to sell you batteries.

If you run your vehicle at a spread of -20F to 120F, absolutely not.

I can promise you that the vehicles we are working on go past 2032 and they’re all still ICE.

These EVa are just not replacements for ICE. They make great supplements and town cars. It’s a marketing failure, and I’m not sure the pie in the sky “your car will fail before your battery” claim is helpful.

I’m pretty sure the only people that would jump on that are EV owners, which I am, and don’t.

LeFantome 9 days ago [-]
Not sure what kind of EV you have. Ours goes over 150 km most weekdays, somewhere beteeen 20 km and 200 km on weekend days, and has taken the family on multiple 1000 km or more road trips now.

My wife drives the EV most days. I cannot count the number of times I have been running late and had to stop to fill my gas guzzler up. It is freeway most of the way to work ( no gas stations ) and I have risked it more than once. The EV “tank” is full every morning so it never puts her through that.

In practice, I have way more range anxiety in my ICE vehicle than my wife does in her EV.

Supplements and town cars? If I could afford a new vehicle, my ICE vehicle would be the supplement.

dieselgate 9 days ago [-]
You present a valid argument but I think comparing an EV being charged every morning to not always having a full fuel tank in an ICE, in the morning, is apples to oranges. From that perspective the EV has an advantage but it’s not a direct comparison.
hacker_88 9 days ago [-]
You sometimes don't have full tank every morning because that's when you go full up. Going to fill up is a chore much different from plugging it at home .

Charging up at a station is still a bigger chore but on a different relaxed way

thechao 9 days ago [-]
This is my experience. We use the EV for everything, except drives to West Texas — although I saw plenty of EVs in Alpine, and a few down in Terlingua(!). If I had a no-shit 400 mile EV, I'd never look back.
bryanlarsen 9 days ago [-]
Or more chargers in West Texas?

It seems to me that desire for capacity above 250-300 miles is mostly just accommodation for the lack of availability and slowness of chargers.

SV_BubbleTime 7 days ago [-]
> Not sure what kind of EV you have.

Time in/on?

Almost all of them including Ferrari.

audunw 9 days ago [-]
Btw, if you’re talking about models being launched in 2032 or later, then those will almost certainly be very niche models that will have very limited sales.

I made a projection of EV sales in Norway back in 2017, assuming that the growth will follow a logistics curve (S-curve) that most technology transition follows. The sales have tracked that projection almost perfectly ever since.

I crunched the numbers for world EV sales. With a conservative assumption that puts several of the preceding years above the projection curve. This actually gives very similar parameters for the curve as for Norway, but with a time offset of about 7 years. The projection predicts that EV sales will hit 85% market share by 2030.

audunw 9 days ago [-]
EV sales now account for just over 90% of new cars in Norway. This in a country with a winter that’s brutal on range and where people love to drive 2-6 hours to get to their cabin on top of a mountain. I also regularly see people write about their road trips to the other side of Europe on forums.

I’d call that a pretty good demonstration that EVs are a real replacement for ICE at this point for a range of conditions (one exception I will grant you for now is if you have a heavy trailer you intend to drive long distance with)

There’s also not EV models for all niches, mainly because car companies haven’t had time to design them.

Yes, the economics in other countries is not as good and the infrastructure isn’t as good. But that has nothing to do with the capabilities of the vehicle itself.

The claim that the battery of a modern battery could outlast the vehicle and be on the order of 500k miles doesn’t seem at all unreasonable to me. Especially for LiFePo. Our old 2015 Kia Soul EV looks like it will reach about 150k miles on its original battery. But most vehicles now have 2-4 times bigger battery, with better chemistry and much better thermal management. Bigger batteries means it’s much less likely that it will be driven down to below 10%. More likely that the owner will cap charging to 80% to reduce battery degradation and much less use of fast charging. So we’re talking about several layers of non-linear reductions in battery degradation.

We fast charged the old Kia all the time. Our Ioniq 5 we haven’t needed to fast charge a single time.

recursive 9 days ago [-]
My brother in law has a model year 2014 Model S still on the road for daily driving. What's the difference?

I've been using EV exclusively since 2022, including long road trips. EV can't totally replace ICE, but nor can ICE totally replace EV. I can't run an ICE car in a closed environment without poisoning myself to death for instance. But exact replication was never the point.

In addition to great supplements and town cars, they're also great cars. If I can help it, I'll never own another ICE vehicle again.

theshrike79 8 days ago [-]
> These EVa are just not replacements for ICE.*

You need the asterisk there and an explanation that says "for my use".

There are clear and gaping holes in the EV line, mostly for American use-cases where people "need" a F250 crew cab as a commuter and their electric grid will crap out if you plug in too many Christmas lights. And America also has an abundance of people who regularly (As in daily or weekly) commute/travel distances that would take one across 1-3 countries in Europe.

On this side of the big wet bit, there are far fewer holes, but they do exist. There's an abundance of compacts, sedans and normal-sized SUVs - all of which are perfectly enough for day to day use.

Mine does 200-250km comfortably during the warmer months and 150km-ish when it's -20C. The last time I plugged in to a charger other than the one at my house or my parent's was last September when we did a longer road trip as a family and I wasn't sure about the destination hotel's charging situation.

If I need to haul insane loads, I can hire a massive van for 10€/hour. Or I can just pay to have stuff delivered.

SV_BubbleTime 8 days ago [-]
Cool story and all, but no, I’m not telling you may use.

I’m telling you no matter how you want to justify the vehicle, or the range anxiety, or the sitting at charging stations for hours - they are not replacements for ICE vehicles. They’re excellent town cars though.

theshrike79 7 days ago [-]
Like I said, most people don't need more than a "town car". The average American commute is 42 miles per day (~68km in regular units).

You can drive that far at -40C with pretty much even the most beat down first gen Nissan Leaf with an original battery.

And if you can charge at home (like you should if you own an EV), the battery will be at 100% every single morning. So you'll spend around 15 seconds plugging the car in the charger when you come home and 7.8 seconds taking it off the charger.

And with all the money you save on gas by using electricity for commuting, you can buy a huge Mad Max style ICE with 500 gallon tanks you can use to drive coast to coast without stopping to refuel once :D

while_true_ 7 days ago [-]
Huh? Literally millions of people worldwide have replaced their ICE vehicles with EVs. Granted, there are some use cases where ICE is the better choice, but no one is seriously suggesting that EVs will replace ICE 100 percent.
SV_BubbleTime 7 days ago [-]
Assume at least 1/2 are people that will learn a lesson in marketing.

It’s been an assumption trap on both sides. The current gen of EV marketing will go down in history books.

bilsbie 9 days ago [-]
Is your critique of this study that they didn’t use real world conditions?
SV_BubbleTime 9 days ago [-]
My critique is that the original claim that the vehicle will fail before the battery is entirely BS. Is that a plain enough way to say it?
theshrike79 8 days ago [-]
The vehicle will be a ship of Theseus situation before the battery fails =)
SV_BubbleTime 8 days ago [-]
Better hope you keep it parked in temperature controlled storage.
r00fus 10 days ago [-]
I fully expect my 2017 EV to last for 15 years without significant degradation[1]. I see the usability of EV batteries like Solar panels - in about 10 years, they get to at worst 90% range, and are probably usable well into 20+ years.

Battery tech is what makes our present and future tech economy possible [2].

[1] I don't baby my EV - always charge to 100%, use fast chargers, don't worry about getting it to 10% or lower. It has lost maybe 5% of it's rated capacity.

[2] with apologies to CEO Morgan from SMAC.

dainiusse 9 days ago [-]
That is unrealistic expectation for 15 years, especially for a 2017 EV.
audunw 9 days ago [-]
Our 8 year old 2015 Kia Soul EV is about 15% degraded (on the worst cell, which could be replaced individually btw). Battery degradation is generally fairly linear between 100% to 70% SoH. So I don’t see why I shouldn’t expect another 8 years on it.

It depends on how much it is driven of course. The mileage is probably a bit below average on that one. I have a short commute to work. We did some road trips in it though.

You gotta take into account that old EVs with small batteries will probably often be used as a second car, and so used less often. That’s the case for us since we got a newer bigger car as our family grew. We don’t take the Kia on road trips anymore and my wife takes the bus to work.

aetch 9 days ago [-]
15 years is completely realistic, I only started having problems with my 2007 Ford Escape Hybrid after 17 years of heavy use and 260k miles. Keep in mind that 2007 was when the 1st generation hybrids were just coming out, too, so technology is much better today.

Don’t forget that there are other parts of a car that degrades as well with time. Eventually most well used cars will have an engine or powertrain issue if they even make it to 15 years to use without being totaled in an accident or simply sold second hand or misused.

fransje26 9 days ago [-]
The 2007 Ford Escape Hybrids were using a lot of shared parts with the Toyota Prius, so the battery pack is definitely not Li-ion. The newer high-density Li-ion packs will not survive 15 years without loosing significant capacity, if they survive at all.
aetch 9 days ago [-]
The escape hybrid had a nickel cadmium battery. I don’t see why a li-ion battery pack wouldn’t be usable for 15 years with good battery management. Both battery and ICE cars have reduced performance and range due to capacity/efficiency degradation so replacing and recycling a battery or engine at 15 years to restore performance is not unreasonable. No car, EV or ICE lasts forever. Likewise, most people want to upgrade to a newer flashier car before 15 years.
fransje26 8 days ago [-]
> I don’t see why a li-ion battery pack wouldn’t be usable for 15 years.

For 2 reasons, primarily. First, Toyota chose nickel-cadmium batteries because of their higher charge-cycle life, at the cost of a much lower energy density (about half that of li-ion). Secondly, they designed their system so that the discharge rate and the discharge level of their battery pack stayed low, thereby maximizing the lifetime of their battery pack, at the cost, once more, of extra weight.

So those two aspects combined mean that you can expect a much longer real-life usage of your battery pack before reduced performance becomes an issue.

I fully agree that a battery pack replacement after 15 years could be considered reasonable (as long as the build quality of the rest of the car warrants it, which is not a given nowadays). But if the mean-time between replacement is around 5 years, then it becomes unreasonable.

satiric 9 days ago [-]
The battery chemistry is completely different; You can't assume similar characteristics.
dainiusse 9 days ago [-]
Because we have plenty of real life examples for cars that have a lot less age.
treffer 9 days ago [-]
Is it? The article lists 2015 as the year where things improved a lot, 2017 is well past that. The numbers are low and even that's inflated due to recalls.

I've seen >>10 year old laptops where the battery is still good enough to go from charger to charger. Just go to ebay and check out 2009 MacBooks. That's ~15 years now.

I don't think this is unrealistic if you can live with the heavier degradation.

whatever1 9 days ago [-]
When they do though then you are stuck with a brick.

In the US there is a federal mandated warranty of 100k miles. But this leaves the used market in limbo (one of the reasons that EVs have horrible residual value).

If the failure rates are indeed so low then where are the insurance products for extended battery warranties? If the numbers are correct it would be a ludicrous business.

DanielHB 9 days ago [-]
I work in the industry and from what I see the "bad" batteries are usually manufacturing problems and they stop working very early in their lifecycle.

The degradation of charge capacity though is almost always completely dependent on usage patterns though, if you trickle-charge instead of fast-charge, don't do a lot of full cycles (100% to empty) and keep them in reasonable temperatures they will lose very little charge capacity. BMS systems in cars have been getting better and regulating the temperature of the batteries even when the car is off though.

Unfortunately the used market doesn't have a reasonable way to measure charge capacity, the only way out is for regulations to force disclosure of current charge capacity and total power cycled through the battery and tools to get that information out. Those two metrics would provide similar information compared to mileage for a gas-powered car.

Also if I had the capital, space and know-how I would be buying every single dirty-cheap used EV I could get my hands on and store the batteries. Soon those batteries are going to be a gold mine for recycling.

Recycling li-ion batteries is like aluminium recycling, very energy intensive but you get almost all the raw materials back and recycled aluminium is cheaper than newly mined aluminium. Soon you will see people shipping those used batteries to places with cheap energy for recycling (like Iceland for aluminium recycling). The process is more complicated than aluminium and there aren't that many spent batteries around which is why this is not a big thing yet.

bilsbie 9 days ago [-]
You’d think you could run a battery test if buying a used Tesla. Isn’t that in the service settings?
DanielHB 9 days ago [-]
It needs to be standardised and illegal to tamper with like tampering with mileage clocks in combustion engine cars. Also how it is calculated needs to be certified by a 3rd party.
audunw 9 days ago [-]
Why do you say that? You can get a new battery for the 2011 Nissan Leaf. You can even upgrade to a bigger battery than the original had without an increase in weight.

It’s expensive, perhaps not worth it to most. But it’s a very, very niche model. And many of them still work just fine on their original battery. The market for battery replacement is tiny. No economies of scale.

Imagine 15-30 year down the line when there are millions of Teslas, Ioniqs, etc in the market for a replacement battery, and we have a huge battery recycling industry willing to pay a decent sum for an old car battery. And significantly reduced costs for producing new cells.

Getting the battery replaced will be just like getting a new transmission for an ICE vehicle. Big investment. Painful on your wallet. But it’ll be worth it. Especially considering that the motor and inverter can probably last another 30 years without any issues.

ZeroGravitas 9 days ago [-]
These products exist. Why do you think they don't? It's a fairly obvious outgrowth of similar extended warranty products for ICE cars, where people would also be stuck with a brick if a major component failed.

A more interesting question is whether the prices reflect this science. Insurance are usually pretty good at this as it's their job, but in small markets they might play it safe.

theshrike79 8 days ago [-]
My car's drivetrain battery has an 8 year warranty. I'm gonna jump with joy if it breaks as fast as people predict. I'll basically get a brand new car for free =)
dieselgate 9 days ago [-]
Cool study and great to see some data on this. The numbers seemed pretty good to start out with - lower single digit failure percentage - but then they get an order of magnitude more reliable to tenths of a percent failure.

I’m mostly curious what happens when an EV has been neglected or left un running for long durations.

metadat 9 days ago [-]
The battery gets depleted in my rivian pretty fast if it's sitting. I left it unplugged when I went out of town for 10 days and came back to a nearly dead truck. It depletes at around 4-8% per day even when parked in the garage and not in active use.
gruturo 9 days ago [-]
I have no idea if this is normal for a Rivian, but it sounds incredibly high.

My Tesla M3 (LFP for the record, I doubt it matters) was parked in my garage for about 25 days while I was abroad and it lost about 2%. Not 2% per day, 2% overall.

Recently I was away for a week and when I came back it was at the exact % I had left it.

bryanlarsen 9 days ago [-]
Sounds like a Tesla with sentry mode on. IOW, there's something in your car that is running 24/7 that is using more than a trickle. Hopefully you can turn it off.
bilsbie 9 days ago [-]
Why does sentry mode need so much power? You’d think 20-30 watts to power the cameras, modem, and a small computer.
bryanlarsen 9 days ago [-]
Sentry mode requires the high voltage system to be on, most of the ~300 watts goes to conversion inefficiency in that system.

(source: the internet, so...)

kybernetikos 9 days ago [-]
That seems wrong, might be worth getting it checked. (I have an EV but not a rivian)
metadat 9 days ago [-]
The technicians say it's working as intended, maybe I have a setting wrong? I'm sure it's doing ..something.. Shrug
jeffrallen 9 days ago [-]
Tell them there is NO WAY that is normal. I have left my Renault Zoe for a month and a half with a few percent loss.

You have a warranty issue. Get it fixed before it's too late.

metadat 9 days ago [-]
A Renault is kinda irrelevant, I have both a Rivian R1T and R1S. They both behave the same.

It's really up to the manufacturer how active they want the system to be when it's sitting in park. I'm sure they could improve it, but as another poster mentioned, maybe it's in a sentry mode or similar.

I have an e-bike that I almost never plug in and it holds its charge just fine with almost no depletion after several months of sitting.

Your mileage will vary :)

bilsbie 9 days ago [-]
I’ve found that annoying in the Tesla too. Why does it need 2kwh to check in with the app once an hour overnight?

I even made sure everything was turned off.

thebruce87m 9 days ago [-]
That seems like a lot still - I’ve left mine parked at the airport for a week and seen no meaningful drain.
r00fus 9 days ago [-]
Completely not my experience - 1 week if I leave mine at 100% it is like 97% when I get back.
jeffrallen 9 days ago [-]
Another data point: the startups Renault was funding to reuse old vehicle batteries in stationary applications once they went below 85% capacity were starved of batteries to use, because the Renault battery packs last so long. They had to pivot to recycling tool batteries by shredding them!

Source: I have a 6 year old Renault Zoe with more than 95% of it's original capacity. The only thing that's wearing out is the suspension and interior. The battery will outlast 3 chassis at this rate, it it was possible to transfer it to a new car.

erictd 9 days ago [-]
Since there are zillions of Model Ys out there, what will the replacement battery market look like when mine finally fails in n more years? Will there be relatively inexpensive batteries with the same kWh as mine? Will there be more expensive batteries with more or much more storage available? I guess I'll find out.
theshrike79 8 days ago [-]
There's already a secondhand market that replaces and upgrades Nissan Leaf batteries.

Same goes for Teslas. The official Tesla service will always replace the whole slab for massive $$$$. Smaller independent shops (mostly self-taught currently) can replace individual failed blocks from donor batteries pretty cheaply. (Less money than an head gasket change or a clutch replacement.)

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact
Rendered at 16:12:44 GMT+0000 (Coordinated Universal Time) with Vercel.