Do you think fast charging has nothing to do with battery degradation?

I recently came across a post on Reddit’s DrEVdev where someone cited an article claiming that Tesla Supercharging is not related to battery degradation. I will not specify the original source here. Collecting and analyzing such data is not an easy task, and I have no intention of criticizing the author.

However, since the article has been widely referenced in blogs and YouTube videos, leading some people to believe that fast charging has no relation to battery degradation, I would like to point out a few issues that deserve attention.

Before discussing the original article itself, let’s briefly look at the background of charging research in the field of battery management.

After reviewing thousands of SCI papers, I have found many experimental studies showing that charging speed and degradation are correlated, but I have never seen a study concluding that they are unrelated. If such a paper exists, I would like to read it carefully.

In engineering, the correlation between charge rate and degradation is considered basic knowledge during the design stage. The degree of impact can vary depending on the charging protocol, such as current, voltage, and temperature conditions.

One of the most active research topics in battery management today is how to minimize degradation while enabling fast charging. If fast charging truly had no effect on degradation, there would be no reason for so many researchers to spend significant time and resources studying ways to reduce its impact.

Tesla’s battery heating function during Supercharging is also based on scientific findings showing that preheating the battery during fast charging can help reduce degradation. Tesla applied this concept directly in its production vehicles. The key point here is not that fast charging is unrelated to degradation, but that Tesla implemented a way to reduce its effects.

Now, let’s look at the article that has been widely cited. The original study compared vehicles that used fast charging less than 30 percent of the time with those that used it more than 70 percent. I will not include the chart here due to copyright concerns, but its title translates roughly to “Fast charging may not accelerate range loss.” The wording “may not” is important; it does not say it does not.

There are several issues with this analysis. None of the key factors that influence battery health were controlled. Under such conditions, it is difficult to consider the results scientifically valid. Another issue is that the study used driving range, an indirect and imprecise indicator, as a proxy for battery degradation. Even so, when you look at the chart, it could actually suggest that fast charging may still have some effect even in an uncontrolled dataset.

As for the sample size, the fast charging group included only 344 vehicles, while the comparison group had 13,059 vehicles. This is statistically very unbalanced, and with so many uncontrolled variables, it is hard to draw any meaningful conclusion from only 344 samples.

In the author’s conclusion, they acknowledge that since the data mostly represent relatively new vehicles, it is too early to determine long term effects, and the article ends by advising readers to avoid fast charging when battery temperature or state of charge is too high or too low.

Below is a graph from Geotab in Australia that visualized the same topic through data analysis, but it shows the opposite trend. Regardless of its absolute reliability, the direction of the result is different.

To overturn an established principle, a rigorous experimental design and strong logical evidence are required. In my personal view, the article seems less like a piece written from scientific conviction and more like one crafted to attract attention through a provocative or marketing oriented title. That is likely why it has been so widely quoted in blogs and YouTube videos. It challenges common understanding.

Lastly, here is a dataset I used in my 2021 research on battery degradation prediction using machine learning. It shows cycle life under different charging protocols, and the difference in battery lifespan varies significantly depending on the charging conditions.

Attia, P. M. et al. Closed-loop optimization of fast-charging protocols for batteries with machine learning. Nature 578, 397–402 (2020).