Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

My Rated Range varies with temp change...

KenC

Active Member
Sep 4, 2018
3,490
3,176
Maine
So, I finally got around to adding the temperature data from the last year to my Stats battery health data. Unlike Teslafi, we don't get the temp data, so I had to add it in manually.

This is my Stats Battery Health data. As I got my car in December, it's now been almost a year. Ignore the straight line, as you can see it's starting to show a curve. My stats data is missing my first 2000 miles, due to software issues, when my rated range was lower, as low as 287 miles. Originally, I charged to 90%, because it was Winter. Then as Spring hit, I changed to 80%, and my Rated Range improved. Based upon that one datapoint, I thought perhaps the more conservative approach to charging improved my range. Now that Winter has hit, my Rated Range has started to drop again, which kind of disproves that theory.

But, it raises another one, that my Rated Range is varying with ambient temperature change.
IMG_4399.jpeg


So, I finally got around to exporting my Stats data, and adding the temperature data manually, and this is what it shows. Luckily, I also had my range data from January/February, all the way back to 400 miles on the odometer.

I've graphed the Temperature on the left axis in red, and the Rated Range on the right axis in blue. I adjusted the scale to get them to overlay, and as you can see, as the seasons pass and the temp changes, my Rated Range bounces in sync. There's a red horizontal line at 310 miles to show my battery is right there during the summer months. And the dark blue squiggly line is a moving average to smooth out the bounces.

So, there seems to be a strong correlation between my Rated Range as reported by Stats and outside temps. My car is outdoors year-round, as I don't have a garage.
Screenshot 2019-12-08 00.36.52.jpg

Now, I think there may be at least 3 things going on, making analysis difficult: one, BMS drift, like mine, due to seasonal temp change; two, sudden BMS drop, which I'm not sure anyone can quite explain yet, could be buffer change, could be software update causing BMS error, etc; and three, real battery deg.

Mine seems to be completely BMS drift. The interesting thing, is if I look at Stats and the Tesla app at the same time, I notice that there's a mismatch when it's cold. Tesla shows my SOC at 59%, while Stats sees it at 60%. If you divide the Rated Range number in the lower left corner, 184miles, by 59%, you get 312 miles; and if by 60%, you get 307 miles, a 5 mile difference. Which is right? Well, the Tesla app is what the car shows, if I toggle the odometer display from range to SOC, it'll show what the Tesla app shows, 184miles and 59%. The Stats app, and the other 3rd party app I have EV Watch, show the same 60%. It lags. And, since it's cold, it always lags high, resulting in Rated Range estimates too low. During the warm months, there's no lag, because the vampire drain is low. During the cold months, the vampire drain is high, so the lag coming from the API becomes apparent. It may be the API doesn't update as often as the car itself does.

My drain in summer was 0.1miles/hr. My drain last winter was 0.4miles/ hr. I would suggest to anyone with possible BMS drift issues, to just toggle their display from miles to SOC and back, and do the math manually. Whenever I get an oddly low Rated Range figure as reported by Stats, when I do the math manually, I always get 308 to 312 miles. My battery isn't showing any change, but if I were to only look at Stats, I'd think there was something going on, when there isn't. Last winter, I vaguely recall, the difference being greater than 1%, as high as 3%. I'll try to capture a screenshot this year.
IMG_4149 (1).jpeg

So, my conclusion is, if you just have BMS drift, like I do, there's nothing wrong with your battery, it'll bounce back when the temps warm up. If you have big drops, like some people I've seen, then I'd consider trying to recalibrate the BMS, once a month, to see if that corrects it. And, if you have true deg, then there's nothing to do, but either try to slow it down, by charging only up to 63%, 3.92V, or speed it up and get 30% deg and file a warranty claim to have your battery replaced.
 

KenC

Active Member
Sep 4, 2018
3,490
3,176
Maine
Adjusted the temp scale a little to see if I could get a better fit between Temperature and Rated Range (estimated), and put in a moving average line for temp. The curves match very well.
Screenshot 2019-12-09 12.51.38.jpg
 

Piwacet

Member
Jun 14, 2019
62
46
SF Bay Area
I've got a hypothesis, based on nothing. So if I'm off or otherwise got this wrong, apologies ahead of time.

Presumably the car can measure fairly accurately the amount of charge (kwh) that goes into the battery when charging. And, presumably, it can measure the voltages of the battery cells through this process.

It may be that the estimator algorithm looks at how much energe it takes to raise the cell voltages a specific amount to estimate the battery capacity.

My assumption is that since a cold battery can not store as much energy, the car sees that it takes less energy to raise the cell voltage a specific amount compared to how much energy it would take in warmer weather to raise the cell voltage the same amount.

In other words (totally made up numbers), if it takes 50 kwh to raise the cells' voltages from 3.6 to 4.1 in warm weather, it may only take 45 kwh to raise the cells voltage from 3.6 to 4.1 in colder weather. If the estimator algorithm does not take the effects of temperature into account, it could perceive this as reduced battery capacity (degradation as opposed to temperature effect). Presumably it's looking at a longer term average than just 1 or a few days, so a warm day in winter would not immediately reverse this effect.

But, I really don't know much about this stuff, this is just speculation.
 
  • Like
Reactions: KenC

AlanSubie4Life

Efficiency Obsessed Member
Oct 22, 2018
9,422
11,318
San Diego
Presumably it's looking at a longer term average than just 1 or a few days, so a warm day in winter would not immediately reverse this effect.

It can be immediate. I saw an obvious loss of range when parked overnight at 35 degrees and it got right back to the original value later that day at 90 degrees. 6-7 mile (2%) change.

In the extreme, a very cold battery gives you a snowflake and blue battery bar and will result in quite a few miles being unavailable until the battery warms up. But they aren’t gone, they are in there - the battery just needs to be warmed up.
 
  • Informative
Reactions: M3BlueGeorgia

Paddy3101

Member
Mar 20, 2019
261
389
San Diego, US
Very interesting, and correlation with temperature does seem like a logical one.

My range drop issue, has halted, and reversed, for no apparent reason (other than maybe software update changes)
Was on a downward trend to about 18K, but gained back nearly 10miles over the last 2K.

I can't put this down to low temperature, as the car is garaged, UNLESS, there is also a negative temp correlation as well when the pack is warmer. The MIN temp I see in the garage, when charging is 58, and the max recorded is 85. (This is overnight from 12-6am approx)

Will take a look at plotting a correlation graph of the data I have. Would make sense that there is an ambient temperature window when charging which gives the 'best' indicated range.

Difference is my rated range, is right at the bottom of your graph. It's just improved UP to 290

upload_2019-12-10_9-24-43.png
 

KenC

Active Member
Sep 4, 2018
3,490
3,176
Maine
Very interesting, and correlation with temperature does seem like a logical one.

My range drop issue, has halted, and reversed, for no apparent reason (other than maybe software update changes)
Was on a downward trend to about 18K, but gained back nearly 10miles over the last 2K.

I can't put this down to low temperature, as the car is garaged, UNLESS, there is also a negative temp correlation as well when the pack is warmer. The MIN temp I see in the garage, when charging is 58, and the max recorded is 85. (This is overnight from 12-6am approx)

Will take a look at plotting a correlation graph of the data I have. Would make sense that there is an ambient temperature window when charging which gives the 'best' indicated range.

Difference is my rated range, is right at the bottom of your graph. It's just improved UP to 290

View attachment 486803
That'll be interesting because most people charge in a garage and not outside in a carport like I do.
 

ivy_

Member
Sep 25, 2019
144
140
Michigan
When I saw this thread this morning, I looked at my rated range on stats. The car had just stopped charging, so the battery was warm, or at least as warm as it's gonna get in December. The garage was around 44°F. The rated range was 245 miles.

Fast forward to now. The car is sitting outside, with a cold soaked battery. The temp is now around 30°F. Rated range is down to 237 miles. I'll see if any comes back after the ride home.
 

KenC

Active Member
Sep 4, 2018
3,490
3,176
Maine
When I saw this thread this morning, I looked at my rated range on stats. The car had just stopped charging, so the battery was warm, or at least as warm as it's gonna get in December. The garage was around 44°F. The rated range was 245 miles.

Fast forward to now. The car is sitting outside, with a cold soaked battery. The temp is now around 30°F. Rated range is down to 237 miles. I'll see if any comes back after the ride home.
Toggle the odometer to see what the SOC shows.
 

jelloslug

Active Member
Jul 21, 2015
4,711
6,018
Greenville, SC
I have been telling people for years here now that the indicated range is not a great indicator of battery health. There are many other variable that are used that make noticeable changes in indicated remaining range. This is one of the reasons that I switched over to percentage of charge quite a while ago.
 
  • Like
Reactions: KenC

ivy_

Member
Sep 25, 2019
144
140
Michigan
When I saw this thread this morning, I looked at my rated range on stats. The car had just stopped charging, so the battery was warm, or at least as warm as it's gonna get in December. The garage was around 44°F. The rated range was 245 miles.

Fast forward to now. The car is sitting outside, with a cold soaked battery. The temp is now around 30°F. Rated range is down to 237 miles. I'll see if any comes back after the ride home.

The drive home raised the range up to 242.
 
  • Like
Reactions: KenC

Jawbone

Member
Jun 10, 2019
167
101
GE
The interesting thing, is if I look at Stats and the Tesla app at the same time, I notice that there's a mismatch when it's cold. Tesla shows my SOC at 59%, while Stats sees it at 60%.

When the battery is cold there will be a diffrence in "battery_level" and "usable_battery_level" as reported by the API. It looks like the stats app returns the battery level, while the Tesla app returns the usable.

Once the battery gets warmer the energy will be available again and both should report the same.
 
  • Informative
Reactions: KenC

ivy_

Member
Sep 25, 2019
144
140
Michigan
When the battery is cold there will be a diffrence in "battery_level" and "usable_battery_level" as reported by the API. It looks like the stats app returns the battery level, while the Tesla app returns the usable.

Once the battery gets warmer the energy will be available again and both should report the same.

Of note, I was reporting the stats value. Might be fun to write a small bit of code to query the api directly.
 

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top