stellantis dodge charger factorial energy hintergrund
Image: Stellantis
BackgroundBattery

New study shows: EV batteries last much longer than expected

While diesel cars run and run, electric car batteries quickly break down and turn the expensive new car into a complete financial loss - statements like these can be found on social media repeatedly. P3 now analysed data from more than 7,000 electric cars, providing insights into real battery ageing. And the field data paints a whole different picture.

Sometimes it’s about child labour in cobalt mining, then it’s about range anxiety or the collapse of the power grid if all electric cars charge at the same time – the internet is full of false statements and half-truths about electric cars and their batteries. As soon as one topic is refuted with facts or made obsolete by longer ranges and better charging networks, the next rumour is spread. One of these myths is battery ageing: used electric cars will be almost impossible to sell as the battery ages quickly, loses range and, in the worst case, has to be replaced – supposedly.

Like so many myths, the one about battery ageing has a kernel of truth – batteries age in two ways. So anyone who wants to find out more about electric cars will sooner or later come across such stories. One thing is clear: the battery is the most expensive component of an electric car, which is why the legitimate question arises as to how this affects the residual value of the car or how often electric car batteries actually need to be replaced and what financial consequences this has.

Management consultancy P3, which specialises in electric mobility, conducted a study to provide a fact-based answer and counter the battery myths. In the first step, P3 examined 50 electric cars from its own fleet and later analysed the real measurement data from 7,000 electric cars. P3 wants to use the results to provide consumers with comprehensive information to eliminate misunderstandings about electric mobility and battery life. “Misinformation can have a negative impact on the transition to electric mobility by fuelling unfounded fears and thus reducing social acceptance and market penetration of electric vehicles. Therefore, providing reliable and transparent data is crucial to provide a realistic picture of actual battery life and thus strengthen confidence in electric vehicles,” the white paper states.

State of health as the main indicator

Before we get to the results, let’s briefly clarify a few terms. The ‘state of health,’ or SoH for short, is central to battery ageing. There is no standardised definition here; in this publication, P3 refers purely to the battery’s capacity. The SoH is defined as the ratio of the current, measured capacity to that in new condition – strictly speaking, the net capacity in each case, i.e. the energy content that the customer can utilise. The gross capacity, i.e., the energy content installed in the vehicle, is higher but irrelevant here. In the end, only the energy available to the customer in the electric car counts. In other words, the net capacity when the battery is new corresponds to a SoH of 100 per cent. If the current capacity drops later, the SoH also drops below 100 per cent. The manufacturer’s warranty conditions often specify 70 or 80 per cent after a certain mileage or period of use.

For completeness, there is often talk of “calendar” and “cyclical” ageing. During calendar ageing of a battery, chemical structures in the battery cells change, even without active use. Cyclical ageing causes additional stress due to the charging and discharging of the battery. Both factors cannot be avoided entirely (more on this later) and cannot be clearly separated from each other. For this reason, the SoH definition does not include values such as the charging and discharging history, but only the capacity – which is relevant for electric cars from the customer’s point of view.

Data from over 7,000 vehicles analysed

What is new about this study is the fact that it is based on real data from vehicles on the road. In its SoH model, P3 previously made predictions about service life based on academic data and laboratory measurements – usually at cell level. The data from the electric cars adds not only external factors such as environmental influences and driving and charging behaviour but also the programming of the battery management system and the ageing strategies implemented by the car manufacturers in terms of how they stress and/or protect their batteries.

P3 took two approaches to obtain the all-important vehicle data: Firstly, 50 vehicles from the company’s own fleet were measured, the battery was examined for previous ageing, and the SoH was correlated with usage and charging behaviour. “The vehicles were selected to gain insights into as many different driving and charging profiles as possible and to identify differences between the manufacturers,” says P3.

To back up these qualitative individual analyses with more quantitative data, the second step involved the Austrian battery diagnostics startup Aviloo. According to the latter, it has already carried out over 60,000 capacity tests. A brief digression: Aviloo offers two different analyses, the ‘Flash Test’ and the ‘Premium Test.’ In both cases, the so-called Aviloo Box, an OBD dongle, is connected to the vehicle’s OBD port. For the ‘Premium Test,’ the battery is charged to 100 per cent and run down to ten per cent with the dongle connected. That way, countless battery-relevant data is measured and analysed on the Aviloo servers. The ‘Flash Test’ only takes place when the vehicle is stationary, and analyses the available measured values based on the database that was set up for the respective battery type via the ‘Premium Test’.

For its analysis, P3 used the data records of over 7,000 vehicles, all of which were analysed using the ‘Premium Test’ – this procedure is more complex but also more precise. This data set also included vehicles with a mileage of over 300,000 kilometres. “This allowed the battery ageing to be assessed in more detail and more quantitatively in relation to mileage,” the white paper states. “This specifically complements the analysis of the P3 fleet and provides a broader database for a well-founded assessment.”

p3 soh whitepaper 2024 01
Image: P3 Group

And the evaluation has a clear result: in the first 30,000 kilometres or so, the loss of capacity is accelerated, meaning that the state of health drops relatively quickly from 100 to around 95 per cent. The good news is that the real degradation decreases with increasing kilometre mileage. The Aviloo data from the 7,000 vehicles shows an (averaged) SoH of around 90 per cent at 100,000 kilometres. And after that, the trend line is almost horizontal; between 200,000 and 300,000 kilometres, it is almost stable – and is well above the 70 to 80 per cent from the battery warranty. In fact, it is closer to 87 per cent.

There is a simple explanation for the rapid SoH loss in the initial phase: a so-called SEI layer (solid electrolyte interphase) forms on the anode (i.e. the negative terminal) in the battery cell during the first charging and discharging cycles. These are deposits of reaction products from the electrolyte that always form. Depending on the vehicle and battery chemistry, this can occur in very different ways, which is why there is a wide variation in the data. However, the trend line from the more than 7,000 vehicle data sets provides a good estimate.

And the data from the 50 P3 vehicles also matches the results from the Aviloo analysis. Some of these vehicles were also used for other tests, so the driving and charging profile of these vehicles may well be more extreme than that of a company car that is primarily used for commuting. Nevertheless, their batteries have proven long-lasting: “Almost all of the P3 vehicles tested have a SoH of over 90%. It indicates that the batteries in the P3 fleet continue to perform very well despite different manufacturers, different usage profiles and intensive use.”

Another interesting finding from the more than 7,000 data sets: The field data suggests that the actual battery capacity is maintained longer than assumed under real-life conditions, especially with the often-cited high mileages of 200,000 kilometres and more. Based on the cell lab tests, the SoH model published by P3 in 2023 gave a much more pessimistic forecast for battery health. Up to around 50,000 kilometres, the laboratory model and the field data are roughly the same – above 100,000 kilometres. However, the trend lines diverge significantly. P3 concludes that the actual user-profiles and the control of the cells by the battery management system in the field significantly reduce ageing.

But how can the observed variation be explained? After all, some vehicles still have an extremely high SoH after more than 50,000 kilometres, while individual vehicles are still at 98 per cent after almost 200,000 kilometres – while others quickly fall below 90 per cent. In fact, the charging and usage behaviour of drivers and the vehicles themselves influence this, as do the manufacturers. On the one hand, the intended buffer (i.e., the difference between gross and net capacity) plays an important role in terms of size and utilisation of the buffer. That is because it can be used, for example, to reduce the noticeable ageing during the warranty period – by releasing a little more net capacity over time. On the other hand, the charging behaviour can be adjusted via a software update. On the one hand, this can be a higher charging power for shorter charging times, which leads to more stress in the cell. On the other hand, it is also possible that an update improves the control of the cells, for example, by optimising preconditioning to reduce stress during fast charging under sub-optimal conditions.

Database deteriorates with higher mileage

Two points of criticism of the data sets should not go unmentioned: P3 itself points out that the database for vehicles with over 200,000 kilometres is significantly smaller than for vehicles with lower mileage. “The reason is that there are only a few vehicles with such a long range. It somewhat limits the data’s validity for high kilometre readings and also leads to greater scattering of the data,” the study says. The “survivorship bias” must also be taken into account. After all, only high-mileage vehicles still roadworthy at 200,000 or 300,000 kilometres were measured. Vehicles no longer in use due to battery failure are not included. That may make the reliability of the vehicles appear too positive. But the big problem is that even if an electric car fails prematurely, according to the ADAC breakdown statistics from 2023, this is only occasionally due to the traction battery. “Individual cases, such as failures caused by special usage behaviour or production errors, can still occur and often occur within the warranty period and therefore rarely pose a financial risk to consumers,” P3 writes.

So, what can consumers do to improve battery health and slow down the ageing process? Roughly speaking, with careful driving and charging behaviour. A distinction must be made between calendar and cyclical ageing for a more detailed answer. Important: These are general statements. In individual cases, different behaviour is possible depending on the vehicle and battery. However, the following P3 recommendations will not damage the battery.

In the case of calendar ageing over time, the main factors are temperature and state of charge. When not in use, batteries prefer low to medium temperatures below 25 degrees, according to P3. Too high a temperature (over 60 degrees is mentioned) is a “driving force for chemical reactions, which leads to accelerated degradation of the capacity.” However, the vehicle or battery management can also help here, as our technical deep dive into the VW Group’s PPE platform shows. The charge level at which an electric car is parked over a longer period of time is also important. A higher charge level means a higher voltage in the cell – which accelerates ageing over a longer period of time. P3 recommends parking the vehicle with a low to medium charge level (ten to 50 per cent) for very long parking periods.

Gentle driving and charging behaviour help the battery

Temperature also plays a role in cyclical ageing, i.e. utilisation, but in a different area. If the battery is used, it should be neither too hot nor too cold. That applies to (fast) charging and driving. High currents (fast charging, heavy acceleration, driving at high speed) are generally not conducive to SoH, but especially at extreme temperatures. In other words, moderate driving behaviour with constant, low speeds and infrequent fast charging at medium temperatures would be ideal – and with a low depth of discharge, i.e. if the charge level remains between 20 and 80 per cent. It is not the end of the world if you deviate from this in individual cases (because fast charging is also required along the motorway in winter). However, a predominantly gentle driving and charging behaviour can slow down battery ageing.

The survey also shows that real battery ageing rarely breaches the terms of the battery warranty. The standard warranty for EV battery systems is currently over eight years or 160,000 kilometres, with the first manufacturers going up to 250,000 kilometres and ten years. Lexus even offers a warranty extension to one million kilometres or ten years on the (admittedly not widely used) UX300e. The battery thus generally lasts much longer than the warranty or other parts of the vehicle.

p3 soh whitepaper 2024 02
Image: P3 Group

With the long service life, a second use of the battery after the ‘first life’ in the vehicle is, of course, also possible, for example, as stationary energy storage – the ‘second life.’ Only after this second use, i.e. around 20 years or more, does a battery go into recycling. At least that’s the model.

To answer the question posed at the beginning about the residual value, it depends heavily on the current utilisation phase. Within the manufacturer’s warranty period, it is naturally higher than after the warranty has expired, even if the battery in the vehicle is still functional. The residual value decreases as it is no longer replaced or repaired under warranty. “Within the warranty period, the loss in value is heavily dependent on ageing and the remaining capacity,” P3 writes. “After the warranty expires, a greater loss in value is to be expected. At the end of the first lifetime, depending on the cost of new batteries, the battery may still have a significant residual value through a second use.”

p3-group.com (white paper for download)

26 Comments

about „New study shows: EV batteries last much longer than expected“
Richard Percival
24.11.2024 um 22:13
Really interesting read. The list of inaccurate comments often made about EVs was particularly pertinent, and reflects a degree of confirmation bias among sceptics that does no one any favours . We need more of this type of analysis being reported by the big media channels. My own experience with an electric van very much fits the data being reported. Although it is relatively low mileage, it still reports a capacity of 90% of its original value, after 3 and a half years. Of course this is just the vehicle's own data, but from a user's point of view it is very encouraging. Battery technology is also continually improving and hopefully we will continue to see myths busted and fears allayed.
Ian Hill
25.11.2024 um 11:20
Only 7000 out of 2 million so a very limited supply limit
Tom Sjolund
27.11.2024 um 10:54
So what, it is a good sample size and gave good data.
Peter Gitta
24.11.2024 um 23:54
I still think they're garbage
Jeff Kidden
26.11.2024 um 06:24
Why do EV haters always seem to be not just stupid, but stupid and angry?
Karl arman
25.11.2024 um 18:33
You think that way because you never owned one. Obviously, what you don’t know about EV’s, is a lot.
Elvinas
25.11.2024 um 07:11
Hm... If vehicle is considered EoL after 12years/300000km and is destined for scraping, then this is bad from environmental perspective as in less rich countries this is considered a quite "new car". And more than half of second market deals are 10-15 years old vehicles. I myself bough almost new 6 years car and now it is approaching 10 years and I would consider using it for next 2-3 years before selling if for second time. Previous cars were run up until 16 and 18 years old my me. Although I do not drive much so I have reached just up to 340-360 thousand. But I know quite a few, who have reached 400 or 500 thousand and car is still in use. Yes they are old ICE cars but if new EV's would not be able to reach that we will still be wasting energy and resources to produce "single use" cars and cause lots of unnecessary pollution.
Tom Sjolund
27.11.2024 um 10:59
"Cause more pollution" oh come on! Older car become more and more polluting with age.
Alexander Terrell
27.11.2024 um 10:16
In the UK, the average age of a scrapped vehicle is 14 years, and should have approximately 100,000 miles on the clock. Normally a vehicle is scrapped when the cost of a repair is greater than the estimated vehicle value. (We recently decided to spend €2400 on a replacement turbo unit for our 15 year old car, but it was a close decision). EVs should be less susceptible to such binary failures. A degradation of the battery means the vehicle can have a new life as e.g. a city run-around. I could imagine a 30 year old Tesla with.a range of "only" 100 miles being used for school runs and shopping. Whereas your 30 year old ICE car just won't start, will fail the inspection, and have lost its exhaust. What will the car industry do if EVs last twice as long?
Steve Hill
26.11.2024 um 09:00
There is not reason why EVs should not last much longer. There is far less to go wrong compared to ICE cars.The problem (identified in the article) is that because EVs are still relatively new, we do not have a lot of 300,000 mile cars from which to extract data yet.Also, older cars had poorer battery technology than we have now, so data based on say early Chevy Bolts and Nissan Leafs would not be representative of the modern EV industry.It would have been good if the study had separated lithium ion and lithium iron phosphate battery types. The latter are increasingly being used in the industry and they retain charge better, and for longer.
Jane
26.11.2024 um 05:49
The picture clearly says "may occur later" for the End of First Life. Btw, the "First Life" data is based on the life of the average internal combustion engine vehicle, although some last well beyond 300,000km, most do not. The article also makes it clear that there are currently very few EV's in the 200,000+km range - this is not because they have failed but because most cars are very young. Most of the cars in the 200,000+km range have either done lots of long distance traveller or have been used as Uber or equivalent (ie lots of rapid charging) and are still looking healthy
Mr Liberty
25.11.2024 um 15:28
EVs battery packs last over 300,000 miles and then continue to work with reduced range. Batteries are basically very high grade ore. It's easier and much less expensive to recycle batteries than to mine refine and process the raw materials from the ground. FYI old car batteries are first used in stationery storage for a decade or more before finally being recycled. Currently they recover just north of 97% of the battery materials.
Gabriel Iordache
25.11.2024 um 10:49
Battery technology and costs came down significantly in the last decade. Replacing s battery is now cheaper than replacing a gas engine. And the new battery is likely larger at same weight.
Paul M
25.11.2024 um 12:58
I have a 2012 chevy volt PHEV. At 12 years old the battery capacity is close to new. Around here where salt is used on winter roads cars don't last much past 15 or 20 years. I have no dought that EVs will last as long as ICE cars. And as the article mentioned the batteries can be used for energy storage after the capacity is reduced in the car. Replacing tires and brakes will be more of a concern, it is for me.
Paul Fargotstein
27.11.2024 um 21:06
With braking regen, the brakes should actually last longer then ICE cars. Assuming of course, you aren't driving like a maniac. We got 150K miles out of a set of front pads on a Prius we had, primarily due to regen. Our Chevy Volt has half the pad left with 80K miles. It has a paddle on the steering wheel to apply regen without engaging the brake pads and for normal driving, does most of a stop.
Arnel
20.12.2024 um 08:07
I have a 2019 Hyundai Kona Ultimate & replaced my brake pads at 246000 kms
John Dixon
25.11.2024 um 13:57
And there is still no reliable and large scale way to dispose of them safely, nor recycle them in a cost effective manner. Great for the environment with the mining of materials, waste products produced and lack or reuse and recycle ability.
Mr Liberty
25.11.2024 um 15:32
Bullshit! Look up Redwood Materials, they currently recycle batteries recovering just over 97% of the battery materials. Batteries are basically very high grade ore saving a lot of money and eventually replacing almost all mining.
John W
25.11.2024 um 14:38
I think truly comprehensive data would have included information/details on batteries that failed. The paper was explicitly performed to show batteries lasted longer than common perception, and then everything that didn't fit the premise was excluded. The results are biased. Some batteries will still fail early and be cost prohibitive for those that need cheap cars, relative to an engine. Typical "old" ICE cars can get a "cheap" used engine or parts installed and go on their merry way. Cheap used batteries or parts to somehow fix them won't be so much of a thing.
Steve Hill
26.11.2024 um 09:06
Not so. There are already companies specialising in reconditioning batteries taken from old or scrapped cars, and offering them as replacements at a deep discount.
Jim Ray
25.11.2024 um 21:55
Tell that to my 2014 Fusion Energy parked in my driveway with a dead battery pack an estimate of $12,000 from the dealership and no concern from Ford in fact the dealership wants to charge me 10 grand for a part I understand they pay $6,600 for
Raoul O'Reilly
25.11.2024 um 23:58
When EV 0 to 100% charge time equals the same three minutes that my ICE car delivers, let's talk.
JT
26.11.2024 um 19:12
When you can take 10 seconds to plug your ICE into a pump at your house so it's fully fueled by morning, with no harmful spills or fumes, at a fraction of the price at the gas station, then let's talk.
John
26.11.2024 um 00:00
If you're in a 10 mile per hour fender bender, the insurance companies scrap the car. Nobody can verify the integrity of the batteries. If they catch on fire, the fire department ropes it off until the fire pits itself out. This is a design defect. Every car that gets scrapped out for this needs to be billed directly to the OEMs.
Michel BAYINGS
26.11.2024 um 08:20
In my Tesla Model S from 2013, I have since almost 3 years a new battery but only got a guarantee on it of max 80.000km or 4 years... Problem a bit with replacing batteries is that they can tell and sell you everything, you never know what they really installed, even if it is at a real Tesla service center.
Alex Johns
03.12.2024 um 19:57
When a manufacturer's battery warranty expires, there are extended warranties available from Altelium up to 12 years old or 320,000 km. If you have a Tesla you can buy a test and extended warranty on the battery and powertrain directly at www.welovebatteries.com

Leave a Reply

Your email address will not be published. Required fields are marked *