[ekr]

Although the thought of getting an electric car has passed through my mind on a few occasions, I'm not 100% familiar with the intricate technical details. (for some reason, the tax incentives where I live are still in favor of continuing with the small petrol car I have. Taxes are primarily a function of weight in the Netherlands, and anything besides a lightweight Dacia Spring would imply significantly higher monthly expenditure for me).

What I'm wondering w.r.t. this article is: wouldn't such fast charging shorten the battery lifespan?

I have experience with ebike batteries. Bosch in particular, with very decent 29E samsung cells, that after 70k km or so, basically halved their capacity. I imagine this effect is severily reduced with a car battery because there are a lot more than 10p, so all the wear is distributed more evenly, and 29E are very old technology.

[pepperoni_pizza]

Research on the current EVs shows that they degrade by on average 2.3 % of original capacity a year, but there is a strong dependency on how much the vehicle is used and how often it is DC fast charged, i.e. there is time based degradation and usage based degradation.

Low use vehicles have degradation of 1.5 % a year, heavily used vehicles mostly slow charged had degradation of 2.2 % a year and heavily used vehicles mostly fast charged had highest degradation of 3 % a year.

Now before you think that means the capacity will halve in X years (33, 23 and 17), the article also notes that the degradation is not linear and it was faster in new vehicles and then slowed down - with no way to know if it will slow down further or continue in this manner, etc, until we have a sufficient sample of 20 years old modern EVs.

Link to article https://www.geotab.com/blog/ev-battery-health/

[magicalhippo]

> the article also notes that the degradation is not linear

Battery degradation is affected by multiple processes which respond primarily to different factors, such as heat, cycles or just time.

This[1] paper goes into some detail about that, and also notes that the typical way batteries were evaluated for longevity in the lab significantly overestimates the degradation compared to batteries used with real-world dynamic loads.

Some quotes:

We found that dynamic cycling enhances battery lifetime by up to 38%. Moreover, we determined the window for the tip-over C-rate that balances time-induced ageing and cycling ageing for this commercially relevant chemistry to be approximately between 0.3C and 0.5C, in the range of realistic average C-rates.

Figure 4a,b illustrates that the degradation is initially dominated by the loss of lithium inventory (QLi). [...] However, as the batteries age, additional degradation mechanisms become important. On the one hand, the positive electrode capacity loss dominates and is impacted by the rest fraction at high SOC [...]

On the other hand, the negative electrode capacity loss (which is less than the positive electrode capacity loss) is impacted by the DoD [...]

Figure 4d shows that, in particular at low average C-rates, when the DoD is beyond 85%, the negative electrode capacity degrades more rapidly, while cells avoiding deep discharge have more preserved negative electrode capacities, in agreement with Fig. 3e. In addition, the DoD has no impact on the positive electrode capacity (Fig. 4e).

SOC = State Of Charge

DoD = Depth of Discharge

[1]: https://www.nature.com/articles/s41560-024-01675-8

[rossjudson]

Where does "20 years" come from? What's wrong with "10 years"?

At the 200,000 mile mark battery life is expected to be ~85%. That's what actual data shows. 200,000 is 13 years of driving 15,000 miles a year.

https://recharged.com/articles/tesla-model-y-battery-degrada...

[pepperoni_pizza]

I picked 20 years arbitrarily, what I meant is that we don't have data on how modern EV batteries will look when 20 years old, because they have not been around that long.

The whole LFP chemistry is pretty new, on automotive timescales, and lot of the older data on degradation comes from the first few generations of Nissan Leaf, which did not have battery heating and cooling.

[therealdrag0]

Rule of thumb for modern EVs is to not care about the battery at all. They are expected to last longer than the car. I would doubt even faster charging significantly changes this, or if it does it’s worth the trade off to those who need it.

[thephyber]

“Rule of thumb” is a heuristic, which is necessarily inaccurate.

My only EV was a 1Gen Nissan Leaf, which is a perfect example of the EV that violates your assumptions.

[twodave]

I don’t think the first gen leaf is what parent had in mind when referring to “modern EVs”…

[thephyber]

Which is why I commented — because there was a blind spot to their point.

I interpret “modern EV” as an EV in the 2010+ era (as opposed to the original EVs from the 1880s-1910s, which were not modern) which were made for streets / commuting (as opposed to golf carts / theme park cars, which have been around for many decades). And I don’t think I’m alone when using this framing.

[therealdrag0]

1. A lot has changed in 15 years.

2. Old leaf batteries are popular with the modding community. I have friends who who’ve done ICE to EV car conversions using old leaf batteries that still work fine.

[rossjudson]

+1000. Who cares. It's good enough.

[bdangubic]

dont say that or used EV prices are gonna skyrocket :)

[Toutouxc]

I believe you meant to write “10S” instead of 10p. I’m not 100% sure, but you were talking about e-bike batteries, which are often 36V, made out of 10 cells (or banks of cells) in series. The nominal voltage of most lithium chemistries is 3.6-3.7V.

EV batteries have many more cells in series, for example my car is 104S, and 800V cars have (obviously) more than 200 cells in series.

And the longevity of car batteries isn’t about wear being distributed “evenly” (a healthy battery can’t really wear “unevenly”, you always load all cells at once). EVs take care of their batteries, they cool them, heat them, balance them periodically, and they don’t actually pull that much power from them. They also keep the cells within pretty conservative voltage limits.

[ekr]

Indeed, I meant 10S. And what I meant by load being distributed along more cells, is that since you have many more cells, current drawn from each is lower. Which greatly prolongs the lifetime.

And hence the question I had with charging too fast. Since discharging faster clearly wears them more quickly, surely charging faster has a similar effect, since it's mostly the reversed process? A question probably easily answered with a query to a LLM.

[Toutouxc]

“Number of cells” doesn’t really tell you anything about current and how it will affect the battery. The number of cells in series gives you the nominal voltage of the entire battery, and the P number (number of cells in parallel) rarely tells you anything useful — three 2000 mAh cells in parallel are equivalent to one 6000 mah cell, and both approaches are valid and used.

What you care about is actually the mass of the cells, basically the total weight of the active material. More material means higher capacity and can withstand more current.

For example, my car is 104S and that’s it, no parallel connections, but the individual cells are huge (~170 Ah each).