Tesla Chronicles #38 - Charging Behaviors - Part 1

One of the most frequently asked question or questions by new EV owners on Facebook EV groups always seem to pertain to charging, and more specifically, what should an EV be charged to on the regular, and should they be charged every day.

Well, about 30 Days ago, I decided to start an experiment. For the first 35,000 miles / 21 months of ownership of my 2021 Tesla Model Y Long Range, my charging habits were mostly just random. I would charge mostly only when my SOC at the end of a day would go below 20-50%, and I would charge up to 80-90% most of the time and 100% when we had a trip planned. Well, I am here to report some interesting results with this first 30 days of testing.

But first, let me describe the experiment. My plan is to attempt to determine if there is any observable change to the degradation of my battery if I change my charging habits. Ideally, I should probably perform this sort of experiment at the same time, with multiple Model Y's, over a long period of time (several months?). Unfortunately, that just isn't possible or feasible at this point, and I want some results in weeks, not months or a year or more. My thought for this initial test is to perform 3 different, 5 week long, testing periods this spring while temps here in SE PA are mostly consistent. Charging will take place in my garage to help normalize the ambient temperatures while the car is charging and charging will be scheduled in accordance with my Time of Use electricity Rate Plan from PECO such that charging will finish before 6am each morning (it actually finished by 5:45am on most mornings). For my first 5 week period, I will charge up to 90% every night, and utilize the data generated by TeslaFi.com's "Battery Report" page. Once this period is over, I plan to repeat this test for 5 weeks, but only charging up to 80% each night, and then finally, I'll return to my previous behavior of just random charging the vehicle and see if my previous degradation trends return or not.

As of the date of this initial report (4/6), I have recorded 31consecutive charges up to 90%, and I am already seeing interesting results that I would like to report, so let's get into it.

First, let's show you the overall degradation of my battery from my first week of ownership until now in the plot below. The first thing you see is that the degradation that occurred over the first 10 months of ownership was pretty aggressive. My vehicle range dropped from 326 down to about 310 miles during this period of time. This seems to be fairly consistent with I have seen most owners have reported. I seemed to experience a gain of about 4 miles of range with the installation of SW version 2022.12.3.2. As such, I have broken the trendline generated for my data so far into two different periods as noted by the Blue and Green data below. Starting at about 0.9 years of ownership in the second grouping of data, I included what a 1% and 3% degradation would look like from that point forward. My current degradation is slightly less than 2% per year.

This next plot is just showing a zoomed in view of the data from roughly the past year. In this plot, I have marked the data recorded since I have initiated this first round of testing where I charged up to 90% every day.

And this last plot is my very preliminary look at the 90% charging results. It is honestly too early to say too much about this, but, with 30 days or so of charging to 90%, it looks like I have not only stopped my battery degradation, but that it has actually started to creep upward and I have gained 1 mile of range back with this charging approach. I have also noticed that the day to day variation (standard deviation) of what Teslafi is reporting as my max range has reduced significantly. We'll see if this continues as this little experiment continues. On Monday, I'll probably switch over to charging up to 80% every night to see if the trend continues or is impacted at all by constant charging at a lower SOC each night.

The Teslafi data/chart is below. You can see the daily variation in the data and how my range degradation has really flattened out toward the end.