The impact of charging on an EV battery’s state of health (SOH) is complex and difficult to evaluate. A new study by Geotab sheds light on the topic of how quickly an EV battery loses capacity.

Geotab analyzed over 22,700 electric vehicles across 21 models using real-world telematics data. The results are very interesting and indicate some general trends.

Overall Longevity

The first takeaway is that EV batteries are “robust and built to last beyond a typical vehicle’s service life”.

The battery capacity decreases over time and with use, though the rate of decrease depends on many factors. In most cases, after eight years of service, users should still have more than 80% of their original capacity.

The average battery capacity degradation was 2.3% per year, suggesting an 81.6% SOH after eight years.

That’s in line with Geotab’s first major study in 2020, which also returned an average of 2.3% (a separate 2023 study showed 1.8% per year).

Geotab notes that EV batteries tend to see an initial, sharper drop-off during the first year or two, when battery capacity can decrease noticeably. After that initial period, the rate of decrease is slower.

Various factors might accelerate or slow down the capacity loss, so let’s take a look at some of the results.

DC Fast Charging vs. AC Charging

In Geotab’s study, the frequency of DC fast charging and charging power are major factors in accelerated EV battery capacity fade.

Frequency: AC vs. DC

The company reports that more frequent DC fast charging (DCFC) accelerates battery degradation. For comparison, Geotab uses two groups of EVs: in the first, the DC fast charging sessions’ share was lower than 12%, while in the second, it was higher than 12%.

• Low Frequency (<12% DCFC): 1.5% per year
• High Frequency (>12% DCFC): 2.5% per year

It’s a simple metric, but it already indicates that the group of EVs that DC fast charged more often than once every eight sessions saw a higher average battery degradation of 2.5% per year.

DC Fast-Charging Power

In the next step, Geotab divided the EV group that DC fast charges more frequently into two smaller groups based on peak power level. The dividing point was the share of sessions with a peak power of 100+ kW.

The EVs that charged at a peak of 100+ kW at over 40% of their charging sessions had the highest average battery capacity degradation of 3.0% per year, compared to 2.2% for EVs that had less than 40% of their charging sessions at 100+ kW.

• Low Frequency (<12% DCFC): 1.5% per year
• High Frequency (>12% DCFC):
  • Low DC Power (<40% at 100+ kW): 2.2% per year
  • High DC Power (>40% at 100+ kW): 3.0% per year

Charging power proved to be the single largest stressor on the battery.

The results of the study clearly suggest that the best practice is to use DC fast charging stations (especially high-power ones) only when necessary, focusing on daily AC charging at home or work.

The EV group that uses DCFC rarely (<12% sessions) noted an average battery capacity fade of 1.5% per year. At that rate, after eight years, the battery would have 88% of its original capacity. The group that frequently DCFCs at high-power chargers averaged 3.0% per year, resulting in twice the battery degradation. After eight years, the battery would be at 76% SOH.

Climate / Temperature

The climate plays a noticeable role in EV battery degradation, though not as much as charging habits, Geotab noted.

According to the study, EVs that operated more than 35% of days above 25°C (77°F) noted a 0.4% higher average annual battery degradation than EVs in colder climates (share <35%).

• Mild climate group: less than 35% of days above 25°C (77°F)
• Hot climate group: more than 35% of days above 25°C (77°F)

The effect of 0.4% per year over a decade can accumulate to a few percent of lower capacity and driving range, on average.

We assume that onboard battery thermal management systems are crucial to limit the impact of climate and overall battery temperatures.

State of Charge

Geotab also tries to answer an important question about keeping EVs at a very high or very low battery state of charge (SOC). We know from many previous reports that, in general, drivers should avoid keeping their EVs at extreme SOC levels (though this depends on additional factors like battery chemistry).

The analysis defines extreme SOC as below 20% or above 80%. As it turns out, only high exposure (over 80% of the total time) to extreme SOC noticeably accelerates battery degradation:

• Low exposure (<50% of total time at extreme SOC): 1.4% per year
• Medium exposure (50%-80% of total time at extreme SOC): 1.5% per year
• High exposure (>80% of total time at extreme SOC): 2.0% per year

That’s an interesting finding, suggesting that everything should be ok even if one keeps an EV at extreme SOC for half of the time. However, an EV sitting fully charged or almost depleted for six days per week might be problematic. Over a decade, the effect would accumulate to several percent of additional battery capacity fade.

Part of the explanation for that conclusion might be the battery buffer, which is usually implemented by manufacturers on both ends (bottom and top). This means that the 0% SOC seen on the display is never a true 0%, and the 100% SOC is never the maximum the battery can take. Thus, we are already not operating on the very extreme.

Utilization

The battery degradation is correlated with utilization. The higher the mileage, the higher the average battery capacity fade.

• Low usage (equivalent of a full charging cycle every 7+ days): 1.5% per year
• Medium usage (equivalent of a full charging cycle every 3-6 days): 1.9% per year
• High usage (equivalent of a full charging cycle every 1-2 days): 2.3% per year

However, when the vehicle operates, it contributes to productivity and revenue (which is especially important for fleets and business use cases), resulting in a higher return on investment. It’s worth maximizing EV usage, despite the fact that it loses battery capacity and range faster.

More Research Needed

Geotab’s study provides valuable insight into the topic of EV battery capacity degradation. Some of the results are a bit surprising or even contradictory to other reports. That’s mostly because of the complexity of the topic.

We would love to see even bigger and more detailed reports, with separate groups for main battery chemistries (NMC, NCA, LFP). There are many factors that impact battery degradation, and each requires a higher resolution to determine its impact on the battery’s state of health.

We can’t forget that some factors interact with each other. For example, a higher charging power will not matter much if a particular EV also has a higher battery capacity. The same concerns an EV with a large battery buffer and an extreme SOC.