Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health

[Steve446]

That was interesting. How did they get so many datapoints? If I'm reading it correctly, and converting from km to miles, my 3 would be the little dot way up by the 499km line, for 310 miles, and around 52,500km, since I'm at 33,000 miles. I guess my 3 is a unicorn. According to Stats' gaussian distribution my 3 is only 93rd percentile. This chart makes it look my 3 is 99th+ percentile. The data does show no 3s, in its database, with zero deg beyond my current mileage, so it'll be interesting to see how much longer it can go before that top buffer is exhausted.
View attachment 797540

I think it's an aggregate plot of all M3's (maybe I'm wrong). I certainly found a data point corresponding roughly to my M3. It just confirmed overall what I had already measured with SMT and then calculated from that data. I do acknowledge that the sea of point is quite dense and hard to trace though.

 

[UncertainTimes]

it’s all estimation

 

[GtiMart]

I agree with what was posted. Just note that this is not degradation. The amount of energy currently available in the battery seems to go down, which might be an actual energy loss or an estimation change. Degradation relates to the battery losing some of its ability to store energy in a permanent way.

 

[Lindenwood]

Has anyone played with a M3P with the new battery yet? Here is another post from the referenced thread for context:

 

[DrChaos]

I agree with what was posted. Just note that this is not degradation. The amount of energy currently available in the battery seems to go down, which might be an actual energy loss or an estimation change. Degradation relates to the battery losing some of its ability to store energy in a permanent way.

Question to the battery scientists: what is physically happening when the battery is colder to reduce the energy capacity?

Suppose you charge the battery up to a given SOC when warm, and then it cools off. The range available declines. What happened to the chemical potential energy? Of course there is an increase in internal resistance, meaning that the efficiency is lower, but is there any other mechanism that contributes to true range loss or energy loss?

 

[AAKEE]

The majority of battery degradation occurs in the first 24k miles. You're unlikely to have 195 miles at this point. If you want to recalibrate the BMS, take the car on a long trip with the battery starting around 90%, until the battery is below 50%. Leave it sit there several hours or overnight.

The majority of the degradation occurs on the first years, and the miles driven do not affect this very much.

 

[Paige111]

Are you talking about about SMT? Their instructions for opening the plastic cover are very much correct. Just don't be too forceful.

I dont want to damage my car in any way -I bought the Tesla scan -OBD -but can’t nudge the back area out to get to port area -I dont want to damage my interior area either

 

[Paige111]

The majority of battery degradation occurs in the first 24k miles. You're unlikely to have 195 miles at this point. If you want to recalibrate the BMS, take the car on a long trip with the battery starting around 90%, until the battery is below 50%. Leave it sit there several hours or overnight.

I have 8K miles -7 months -I have a long commute daily, My battery degraded approx 2%, I bought the acceleration boost, but had to return it, I actually saw my biggest range drop w/that software, I was getting approx 353 prior to acceleration boost -it dropped to 347 -even though I recturned the acceleration boost, my total range stays @347

 

[Apprunner]

I have 8K miles -7 months -I have a long commute daily, My battery degraded approx 2%, I bought the acceleration boost, but had to return it, I actually saw my biggest range drop w/that software, I was getting approx 353 prior to acceleration boost -it dropped to 347 -even though I recturned the acceleration boost, my total range stays @347

I'm sorry but the Boost does nothing to impact maximum range. Just a coincidence that the range dropped after the Boost upgrade.

 

[Steve446]

I have 8K miles -7 months -I have a long commute daily, My battery degraded approx 2%, I bought the acceleration boost, but had to return it, I actually saw my biggest range drop w/that software, I was getting approx 353 prior to acceleration boost -it dropped to 347 -even though I recturned the acceleration boost, my total range stays @347

Have you tried Supercharging or attempting a BMS recalibration? Equally, the drop you mention - 347 against 353 corresponds to a 1.3% drop. If you add that to your stated 2% you are still in what's often considered "normal" depending on car use etc. Look at the spread at 11 200 Km ( 7 000 M) on Degradation (yay it's working again!).

 

[Steve446]

I dont want to damage my car in any way -I bought the Tesla scan -OBD -but can’t nudge the back area out to get to port area -I dont want to damage my interior area either

Did you manage to fit the dongle?

 

[FlatSix911]

Sorry, no. I've thought about it, but just haven't had the time to buy and do it. Maybe, if I find more time this Summer.

Well, Stats has offered a new bit of data in the last few months, which shows a comparison of your Tesla's estimated range relative to others, and apparently, I'm at 93%ile, which means there are 7% better than mine, for the same mileage and model. Yes, same mileage and model. I specifically asked the developer if he was also comparing a 2021 to my 2018, and he said he wasn't.
View attachment 796423
As you can see, I'm at 93%ile, even though I'm at 311 miles estimated for a 2018 LR-AWD.
View attachment 796441
Here's the twitter thread where I asked him about other model years, and he again said it was same year and model. Funny thing is that was 2000 miles ago, my range is up a bit to 311 estimated, and yet my percentile is down to 93% from 95%. Whatever. Must be limited datapoints or something.

Interesting data... I still don't understand how your range could increase from 300 miles when new to 310 at 30K miles.

 

[KenC]

Interesting data... I still don't understand how your range could increase from 300 miles when new to 310 at 30K miles.

Ok, at the time when I got my 3, it was December of 2018 and cold. The SOC api that the Stats developer used, was temp-affected. That is, if it was cold, the estimated range would be lower. Do you see those dots in the 280 to 285 mile range? Those were blue snowflake days. That's the tipoff that the developer was using the temp-affected SOC api. I wrote to him to complain. It took until Jan of 2021 before he switched SOC apis, to using the one that isn't temp-affected. In my chart that happened around 14k miles. Notice after that, my range estimate has about 310miles.

So, the 300 miles estimated range when I got my 3, were temp-affected, and not the true estimated range.

 

[FlatSix911]

Ok, at the time when I got my 3, it was December of 2018 and cold. The SOC api that the Stats developer used, was temp-affected. That is, if it was cold, the estimated range would be lower. Do you see those dots in the 280 to 285 mile range? Those were blue snowflake days. That's the tipoff that the developer was using the temp-affected SOC api. I wrote to him to complain. It took until Jan of 2021 before he switched SOC apis, to using the one that isn't temp-affected. In my chart that happened around 14k miles. Notice after that, my range estimate has about 310miles.

So, the 300 miles estimated range when I got my 3, were temp-affected, and not the true estimated range.
View attachment 798649

Thanks for the information on the change in SOC reporting.

 

[jo6829]

Hi

I have an M3 SR+ manufactured 2020 October and bought 2021 April. Ie driven for about 1 year with 16,000 kilometers (10,000 miles) on the odometer.

When I charge it to 100% it estimates 356km (220 miles).

1) I believe it is listed to have 250 miles so does that mean it lost approx 12% capacity already after just 1 year and 10k miles driving?

2) I keep the battery between 10-90% ie not as careful as 20-80% that I read many others are doing, could that be a factor why it lost so much?

3) Can the range improve again if I keep taking care of the battery, eg keeping it between 20-80% from here on?

4) I read that the battery warranty kicks in after 70% capacity loss within 8 years so does that mean that the full charge range has to be as low as approx 175 miles to get the warranty?

 

[Tam]

...When I charge it to 100% it estimates 356km (220 miles)...

Your picture indicates that you set it to 100% but it has not completely charged at 100%. Thus, you described it as "estimates".

You need to let the BMS reaches 100% so it can calculate at that actual level.

That's the missing part. Otherwise you maths are correct.

Enjoy your car. There’s not much you can do until it goes down to 70% of the original specs or 175 miles.

That's why when buying, we need to see what the warranty says. It's great if your car can still charge to 250 miles but the warranty only guarantees it for 175 miles.

Thus, if we want 250 miles, we'd better buy one that can go 360 miles because it guarantees 70% of 360 = 252 miles.

 

[jjrandorin]

I read that the battery warranty kicks in after 70% capacity loss within 8 years so does that mean that the full charge range has to be as low as approx 175 miles to get the warranty?

Yes, 175 miles rated range, as shown by teslas tools, would be required to trigger any degradation claim (roughly 281 km according to an online conversion calculator).

 

[STS-134]

I would like to see the part of the manual stating that the car do not want to stay low in SOC for long time. My guess is that the car manufacturer encourages the car owner to charge at low SOC for the exact same reason that the gas meter is red colored in the low reserve part and there might also be a red lamp on.

Tesla colors the battery amber/yellow > 20% and red below 10%. When parking there is a note comming up on the screen about something like this:
“Weak battery” “Plan your next charge”
Many people takes this as a sign of that the battery will get damaged from low SOC.

The 12v lead acid battery is sensitive to getting discharged but this should only happen after the big battery disconnects due to low SOC. Thos is described below.
View attachment 796739

View attachment 796740

Tesla have no information about low SOC should be bad for the LiB. Theres is information which people have interpreted as “because its bad for the battery” but if you look carefully there is no information saying this, only text that wrongly can be read out of the context and then our minds fill in the blank parts with imagined parts that is not correct.

There is research for most battery types, including Samsung SDI NMC. From what I have seen, there really is not a noticable difference and these behave as expected For cells with NMC-chemistry.
https://arxiv.org/pdf/2007.01937.pdf



I fail to find any information in the article in the link about low SOC being bad. I also ca not find any information about minimum cell stress being in the middle. You need to point out the exact text that include those statements.

As I did write yesterday, we need to understand that there is a big difference between overdischarge batteries that is discharge below the low volt limit set by the manufacturers. For most lithium ion batteries 2.5 to 4.20 V/cell is the allowed range.
0% SOC is when reaching the minimum voltage during discharge.

All research (litterally all) show us that the lower the SOC, within the specified allowed voltage range, the lower the calendar aging.
There is not really any exceptions.

Here you go: https://css.umich.edu/sites/default/files/publication/CSS20-08.pdf

"Low state of charge/overdischarge" is one of the contributing factors to degradation. The mechanisms that cause degradation under these conditions are shown in the picture on page 4.

In general, for any longer trip, I'll try to center the battery around 50% SoC. For example, if I expect a trip to use 60% SoC, I'll charge to 80% so I can arrive back with 20%. If I expect it to use 50%, then the battery gets charged to 75% and I arrive back with 25%. And so on. However, since it seems like higher SoC is worse than lower SoC, I set Scheduled Departure for just before I leave, such that the battery sits at the lower arrival SoC longer than it sits at the higher departure SoC. For local driving, 60% is always used as the charge limit. How am I doing on degradation? Well, after over a year...

 

[AAKEE]

Here you go: https://css.umich.edu/sites/default/files/publication/CSS20-08.pdf

"Low state of charge/overdischarge" is one of the contributing factors to degradation. The mechanisms that cause degradation under these conditions are shown in the picture on page 4.

So that is the Tesla manual?

I know it is not.

Overdischarging, per definition is discharging the cell below the specified minimum voltage. The minimum voltage is 2.5V / cell for the Panasonic NCA cells.

Tesla use 4.5% energy buffer, that means there is 4.5% SOC below 0% On-Screen SOC, and the cell voltage is around 3.0 volts at 0% On Screen SOC.

From all battery science research we know that the lower the SOC, the better.
Calendar aging benefit from low SOC, the lower the better. This is a graph, which is in line with all research:

The smaller the cycle, the better.
And the lower down in SOC, the better.
This is a cycle test from a research report with small cycles, look and understand the conclusion: the lower SOC, the better:

This is Tesla battery info. Some parts have changed, like that Tesla now use the full 4.20V/cell( thats = 100% SOC according to the industry, and now they use 4.5% SOC as the 0%, with a buffer below.
Tesla 2006 battery info

Did you read all in the ’Tesla manual’ you linked to?
Like this:

What we can see is that for all 60% DoD examples, cycling:
0-60% caused the battery to degrade only 3%.
20-80% caused 9% degradation
40-100-% cause 11% degradation.

If low SOC was bad, we should see a distinguished increase in degradation for the 0-60%, as it is cycled all the way to 0%.
But we do not see this, why?

This still do not mean overdischarged as 0% SOC per deginition is when reaching the minimum specified voltage. In the Tesla/Panasonic cells this is 2.5V/cell, and this will be well below both 0% SOC on the screen and below the voltage where the car shuts off.

Also stated in your link:

If we put all the data togheter we see that the lower the SOC, the better. Smaller DoD is also better, even if I didnt enlong this post by provide proof of that.

 

[STS-134]

So that is the Tesla manual?

I know it is not.

Why does it matter if it's the Tesla manual or not? It's an academic publication on what causes Li-ion degradation, and applies to all Li-ion batteries.

Overdischarging, per definition is discharging the cell below the specified minimum voltage. The minimum voltage is 2.5V / cell for the Panasonic NCA cells.

Overdischarging != low SoC. They are related, but low SoC without overdischarging only causes cathode degradation as per that picture. Go into the overdischarge range and you start also getting anode and inactive material degradation. It's obvious that high SoC is even worse than low SoC because high SoC (without overcharging) shows up as a cause for all 3 methods of degradation: anode degradation, cathode degradation, and inactive material degradation. But that does not mean that low SoC, by itself, doesn't cause degradation.

If you want to minimize degradation, keep your battery close to the center of its SoC range, not at the extreme low end. Intuitively, we can sort of understand this as the extreme low SoC range is just outside the "overdischarged" range, and there aren't really any step functions in physics outside of quantum mechanics.