Tesla battery is an important and expensive part of the car. Naturally, owners are very interested in knowing more about how to preserve their batteries and comparing their battery performance with other cars with the same age and model.
I am the developer of Stats for Tesla app which has several tens of thousands of users (many with multiple cars). As a data scientist I am interested in stories that data can tell. In this article, we will analyze the data related to charging and battery capacity to discover facts about Tesla battery preservation and charging habits of users whose battery capacity is above and below average. So, let's get started.
A note about how to read histograms: The X-axis in each histogram represents the quantity described in the title of the histogram. The Y-axis represents the fraction of data associated with the quantity on the X-axis. Wikipedia has a good article about histograms with lots of examples. If you find histograms and charts confusing, do not worry; I add a short sentence after each histograms to interpret what the chart conveys.
Charging Habits
First we look at statistics for SoC (state of charge). The following graph shows the histogram of minimum SoC. This is the SoC after which users start charging their cars.
Observation: Most users start charging the car after SoC reaches ~25%
How about the maximum SoC? The following chart shows the histogram of the charge limit that owners use. Note that this histogram is for cars that do not have an LFP battery. The LFP batteries can charge to 100% more often (To determine if your vehicle has an LFP Battery, navigate to Controls > Software > Additional Vehicle Information).
Observation: Most users set the charge limit to ~90%
It is also interesting to know the statistics of how much charge owners add to the car in each charging session.
Observation: Most users add ~55% each time they charge their car
How Much Do People Supercharge?
Most people charge at home and some people who do not have access to a home charger always supercharge. This histogram shows the distribution of fraction of times that cars supercharge.
Observation: Vast majority of times, users do not supercharge, but there is a bump at the “fraction =1.0” which represents people who almost always supercharge (eg, homes with no access to a charger in the garage)
How Often Do People Charge the Car?
I do not drive much and charge the car about once week. It looks like I am in the minority in that regard:
Observation: Most users charge their cars every two days
Is There a Winning Charging Strategy?
I have always wondered if there is a charging behavior that can minimize battery degradation. For example, should I set the charge limit to 60% and maintain an SoC that is close to 50% as much as possible? Should I let the SoC reach to single digits and then charge the car to 90%?
There are many recommendations that one can find in various Tesla forums (some even attributed to Tesla service center employees). Some sound plausible (eg, charging from 5% to 100% so that the BMS get calibrated). But I wanted the data to show me what the best charging behavior is (if there is one).
For this study, I selected Model 3, AWD (dual motor), 2019 cars because there are more of them and they have been driven for a few years by now. The magenta line shows a polynomial fit to the data (the trend line).
The trend is quite reasonable and the degradation as a function of miles driven is relatively flat (which is good).
Now, let’s analyze the behavior of cars that are above and below the trend line in the previous graph. The blue histogram is for the “least degradation” and the pink one is associated with “most degradation”. As we can see from the following two charts, there is not an obvious difference between the two distributions.
Observation: How much charge users add in each charging session does not seem to affect the battery degradation significantly.
The next two charts provide another way of looking at the data for the same analysis. The top and bottom of each bar in each graph represent the min and max SoC for that car. Again, blue color represents cars with less battery degradation and pink represents higher degradation.
Is Supercharging Harmful to the Battery?
This is another frequently asked question that I also often wonder about. Again Mr. Data probably has something to say about this.
As before, blue represents “lower degradation” and pink is “higher degradation”.
Observation: Supercharging does not seem to have an obvious effect on battery degradation
Quantitatively speaking, we can even compare the 90 percentile of the supercharging fraction for the two cases (lower degradation even used superchargers slightly more):
90 percentile of lower degradation = 0.34
90 percentile of higher degradation = 0.30
Conclusion
I find these results somewhat comforting and liberating because they tell me that:
- I don’t have to be worried about the battery degradation because it is relatively small and does not go down drastically as the car ages.
- I don’t have to be super careful about how I charge my car. Lower SoC and higher charge limit do not seem to affect battery degradation in a significant way
- Supercharging does not cause severe battery degradation
Please keep in mind that these are statistical observations. As with any statistical analysis, there are always outliers and may contradict some or all of these conclusions.
Finally, if you enjoy knowing more about your Tesla and like doing automation and using Siri to interact with you car from iPhone/iPad or Apple Watch or if you are interested in knowing how your car’s current battery capacity compares with other Tesla cars with the same model and year and odometer, you may like my Stats for Tesla app.
