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

[EVEnergy]

Plug your car in every night - it's in the owner's manual to do that. It helps with the balancing of pack voltage.

50-60% is perfect. To maximize battery life, don't run it to the extreme low/high and keep it there for long.

That said, don't fear charging to above 90% when you need to. When I'm on road trips, I'll almost always drop below 20% before I charge back up. Because of the time it takes, I'll charge to 100% or close to it when I leave home on a road trip but while traveling, I stick usually between 80-20%.

you'll see a lot of people obsessing over the battery and honestly I did too when I first got my M3 a few years back. Now, I don't fear or worry about the battery. Enjoy!

 

[DeltaOne]

Thanks DeltaOne.

So, if I plug in my charger at night it sounds like I should set it to stop charging around 50%-60% (unless needing the car for a long range trip).

Is that correct?

Correct.

I charge to 60%. I'm retired, so no daily commute. For just normal day to day stuff, 60% gets me anywhere I need to go...and back. This works well for me.

I don't charge every night, but the other poster is not wrong for telling you to charge every night. If the car is below 45% when I get home...I charge it back up to 60%. And Scheduled Departure, up to 80% or 90%, if it's needed for a special trip.

So, learn to use Scheduled Departure...you'll eventually need it.

I have Sentry Mode set to off when the car is home. The car can sit for days at a time and only lose 1% or 2% (it occasionally wakes up to charge the 12 volt accessory battery).

 

[JetDriver05]

Thank you everyone. This is very helpful.

 

[Matt The Gamer]

So to start, I have already tried the BMS recalibration to attempt to recover any extra percentage/mileage out of the car but to no avail.

I purchased a used 2019 Model 3 SR+ back in March. Unbeknownst to me, the previous owner(s) were either super aggressive on the car or we have collectively bad luck with the battery. About 59k miles on the car and the Technical support has told me that I am at ~83% retention on the battery. My 100% charge state says I get about ~200 miles, but since I rarely charge to 100% and I mainly go from 80-90% -> 10-20% I generally get a maximum usable range of like ~150 miles. That's terrible. And it's terrible for a 3 year old car. Support told me that because it still falls in acceptable ranges, they won't replace it under warranty.

What can I do about this other than either pay for a new car/battery or just sit and wait for the battery to degrade to the point it's replaceable? I cannot charge at home, I live in an apartment, which is why this is more annoying than it probably should be. Since I'm solely reliant on fast/supercharging anyways, and have to fully charge once a week at that, why not just charge the battery as close to 100% as I can when I need to charge and just force the battery to degrade faster to get it replaced? I plan to own the car for years, but having a car with barely 200 miles of full range is a real pain. And the thought of selling in a year or two just to hand off a poor battery to someone else feels wrong.

 

[jjrandorin]

So to start, I have already tried the BMS recalibration to attempt to recover any extra percentage/mileage out of the car but to no avail.

I purchased a used 2019 Model 3 SR+ back in March. Unbeknownst to me, the previous owner(s) were either super aggressive on the car or we have collectively bad luck with the battery. About 59k miles on the car and the Technical support has told me that I am at ~83% retention on the battery. My 100% charge state says I get about ~200 miles, but since I rarely charge to 100% and I mainly go from 80-90% -> 10-20% I generally get a maximum usable range of like ~150 miles. That's terrible. And it's terrible for a 3 year old car. Support told me that because it still falls in acceptable ranges, they won't replace it under warranty.

What can I do about this other than either pay for a new car/battery or just sit and wait for the battery to degrade to the point it's replaceable? I cannot charge at home, I live in an apartment, which is why this is more annoying than it probably should be. Since I'm solely reliant on fast/supercharging anyways, and have to fully charge once a week at that, why not just charge the battery as close to 100% as I can when I need to charge and just force the battery to degrade faster to get it replaced? I plan to own the car for years, but having a car with barely 200 miles of full range is a real pain. And the thought of selling in a year or two just to hand off a poor battery to someone else feels wrong.

Sell it, and pass the problem onto another buyer (like the original owner did). There is nothing else for you to do. Repeat. There is nothing else for you to do. You are not going to buy a battery which is 1/3 the cost of the car, and tesla is not going to replace the battery until it reaches at least 70%.

There is no guarantee if you supercharge it all the time etc it will hit 70% before the warranty period of 100k miles.

There is nothing else for you to do, beside sell the car and buy a LR model 3 (since you are worried about range) or buy some other EV type, or go back to ICE vehicles. The question is usually "how can i get tesla to replace this early, and the blunt answer is, you cant, and they wont, unless it fails or meets the threshhold.. period.

 

[AAKEE]

I did finally get my Panasonic 2170 order delvered yesterday.
25 pieces of ”true model 3” batteries. NCR21700, NCA chemistry.

I did get another 10 pack a few months ago but these most probably was fake, and had too low capacity. The new ones, though….most probably the right type to be able to prove the research data.

I also have two other orders( 4+4cells) to get cells from two other batches. This to make sure that *anyone* can rely on that the data/findings is not only validfor a certain batch of batteries.

I will cycle, test, store and regularly check the capacity and use different SOC for the cells. I will also try to find a few places with higher temperature to see the temperature dependency.

Any specific suggestion for the tests?

 

[AAKEE]

Thank you everyone. This is very helpful.

Actually, there a big win with using 55% or less as the charging target. This is specially valid if you live in an hot area and do not drive the car asap afterthe charges.

55% SOC will more or less half the degradation from time(calendar aging) compared to 60% and above. Going lower than 55% reduce it further.

Also, the low SOC stress is a myth. 0% SOC causes the lowest degradation and 0% on the screen is about 4.5% true SOC, and the battery shut down long beforenthere could be anydanger for the HW battery.

 

[Dave EV]

Plug your car in every night - it's in the owner's manual to do that. It helps with the balancing of pack voltage.

While the manual tells you to plug in every night, that's really only because Tesla wants you to avoid running out of charge and avoid range anxiety. I recommend only plugging in when you need to charge.

Plugging in has nothing to do with "capacity loss anxiety" and plugging in also has nothing to do with balancing the cells in the pack.

The Tesla BMS can balance cells at any SOC and time, but it has the best information from the cells to do so at high SOC (>=90%) because the SOC / Voltage curve of the cells is pretty flat.

Of course, high SOC is detrimental to long-term degradation (See all of AAKEE's posts), but the BMS seems to do a good enough job of keeping things balanced even without charging to high SOC - unless you have a Tesla with LFP cells. Those cells have a very, very flat SOC / Voltage curve so you have to charge to 100% regularly (Tesla recommends once/week, but that's probably conservative and you could probably go less frequently than that - once a month or even less frequently) to give the BMS enough information about individual module SOC so the BMS can balance the pack properly.

End result - if you want to maximize capacity of your pack over time, keep the average SOC as low as reasonable and time your charges using the departure timer or manually so that the car finishes charging right before you drive.

If any of this impedes on the usability of the car, adjust accordingly and don't worry about perfecting things. For example - Performance models need high SOC for maximum power, so if that's what you want, do it and have fun! For the majority of owners, charging to a reasonable SOC < 90% regularly is good enough.

 

[citywok]

Actually, there a big win with using 55% or less as the charging target. This is specially valid if you live in an hot area and do not drive the car asap afterthe charges.

55% SOC will more or less half the degradation from time(calendar aging) compared to 60% and above. Going lower than 55% reduce it further.

Also, the low SOC stress is a myth. 0% SOC causes the lowest degradation and 0% on the screen is about 4.5% true SOC, and the battery shut down long beforenthere could be anydanger for the HW battery.

To keep it "safely" below this threshold of accelerated degradation, shouldn't you store it just a bit lower than 55%?
The true SoC is higher than the on-screen SoC after all.
Assuming the sub-zero 4,5% buffer is equal to 4,5% above zero energy wise, the threshold for NCA/Panasonic would be (57% ÷ 104,5 =) 53,5% and for NMC/LG 59%.
But eh, small differences of course. Could be even different if the SoC scale is not linear, but I believe it is.

 

[AAKEE]

To keep it "safely" below this threshold of accelerated degradation, shouldn't you store it just a bit lower than 55%?
The true SoC is higher than the on-screen SoC after all.

Well, yes. But still no, but yes*

When writing posts, sometimes I feel it could be an good idea to leave some thing out to try to not go so deep into things that the reader stops reading. It probably gets strange for people seeing a step at 55%, but we say that 55% isnt really 55%…

We have been into this before.

True SOC should be:
Screen SOC + (4.5 - 0.045 Screen SOC)

Screen SOC of 55% = 57 % true SOC.
Screen SOC of 53% = 55% true SOC.

*) Theres one caveat. On the same issue that many earlier calendar aging tests had too few datapoints (for example 20 - 40 - 60 - 80 - 100% SOC) and the researchers did incorrectly draw straight lines between these points, the step most certainly do not ”begin” at 55% and end at 60% SOC.
If we had 5 or 10 datapoints between 55 and 60% we would know exactly where the step ”begins”.
The central graphite peak is somewhere around 57% SOC for NCA, if I remember it correct. So it is probably a steeper line starting around 57% rather than 55%.
So, it might / could be / is possible that 57% true SOC , which is 55% screen SOC, is on the right side of the real step.

I charge to 55%, but my car stands most of the time with lower SOC, as the charging is done late in the night.

The lower the SOC, the lower the degradation. 0% will cause less degradation and 0% reduce the calendar aging to almost none. So 53% screen SOC (55% true) is probably better than 55% screen SOC.
On the other hand:
0% beats 10%
10% beats 20%
20% beats 30%
And so on…

 

[FlatSix911]

Tesla releases new software update with better range calculation and media accounts linked to driver profiles

Tesla has started to release a new software update (2022.16.0.2) with a better range calculation incorporating more environmental factors and...

electrek.co

We reported on it back in March when it showed up in the code, but now Tesla appears to be confident enough to put it in the release notes:

“Energy prediction for your route has been improved by incorporating forecasted crosswind, headwind, humidity and ambient temperature when using online navigation.”

 

[Steve446]

Well, yes. But still no, but yes*

When writing posts, sometimes I feel it could be an good idea to leave some thing out to try to not go so deep into things that the reader stops reading. It probably gets strange for people seeing a step at 55%, but we say that 55% isnt really 55%…

We have been into this before.

True SOC should be:
Screen SOC + (4.5 - 0.045 Screen SOC)

Screen SOC of 55% = 57 % true SOC.
Screen SOC of 53% = 55% true SOC.

*) Theres one caveat. On the same issue that many earlier calendar aging tests had too few datapoints (for example 20 - 40 - 60 - 80 - 100% SOC) and the researchers did incorrectly draw straight lines between these points, the step most certainly do not ”begin” at 55% and end at 60% SOC.
If we had 5 or 10 datapoints between 55 and 60% we would know exactly where the step ”begins”.
The central graphite peak is somewhere around 57% SOC for NCA, if I remember it correct. So it is probably a steeper line starting around 57% rather than 55%.
So, it might / could be / is possible that 57% true SOC , which is 55% screen SOC, is on the right side of the real step.

I charge to 55%, but my car stands most of the time with lower SOC, as the charging is done late in the night.

The lower the SOC, the lower the degradation. 0% will cause less degradation and 0% reduce the calendar aging to almost none. So 53% screen SOC (55% true) is probably better than 55% screen SOC.
On the other hand:
0% beats 10%
10% beats 20%
20% beats 30%
And so on…

A good reminder. Concerning this post and your #4886 above, I had a few thoughts about testing:
- how good is the temperature control of the thermal chamber you use - the point I'm getting at is how precise and evenly distributed does the °C control have to be, we see banding with the results (stability @ different temperatures) so how "granular" should that parameter be, in the same way as how "granular" should the definition of SoC be?:
- should you be paying attention to a dynamic temperature profile, after all, a car's battery will likely be in a situation of changing whole battery mean temperature depending on use, time of day, weather and season etc, or can we just band the temperatures and ignore what goes on inside a given band?:
- it would as you said be good to see data for the thermal bands not tested 0, 10, 20 etc, people often quote the Arrhenius relationship so some clearer data would be good particularly with regard to the dynamic temperature profile that actually takes place and likely subtly changes the data or makes it more noisy.
All that takes time, effort and batteries I appreciate

 

[AAKEE]

- how good is the temperature control of the thermal chamber you use - the point I'm getting at is how precise and evenly

I will not use a climate chamber. My main goal is to show that the research is valid for these 2170 cells as well. If I get (Im sure that I will) the same results as the researchers get, theres no reason for anyone to doubt that the research for NCA is valid for the tesla/Panasonic 2170 as well.

Having some cells at slighty higher ambient temp and some at slightly lower probably will confirm the temperature dependent calendar aging.

 

[Dave EV]

Very interesting study from Jeff Dahn et al on how limiting cell voltage on common NMC chemistries to 3.65V will lead to vastly improved lifetime of batteries. and a NMC battery designed for 3.8V max operation beats LFP in energy density and is projected to last nearly a century at room temperature.

Note - the large increase in battery life also should apply to any NCO/NMC cell charged to a lower maximum voltage. 3.65V is right around the 40% SOC mark, I believe and 3.80V is right around 50% (I need to use ScanMyTesla to confirm).

Tesla, drop the minimum charge level to 40%!

Radware Bot Manager Captcha

 

[AAKEE]

I believe and 3.80V is right around 50% (I need to use ScanMyTesla to confirm).

Tesla, drop the minimum charge level to 40%!

3.80 Volt would be close to 53% on the screen. Thats 55% true SOC, by no coincidence.

 

[FlatSix911]

Very interesting study from Jeff Dahn et al on how limiting cell voltage on common NMC chemistries to 3.65V will lead to vastly improved lifetime of batteries. and a NMC battery designed for 3.8V max operation beats LFP in energy density and is projected to last nearly a century at room temperature.

Note - the large increase in battery life also should apply to any NCO/NMC cell charged to a lower maximum voltage. 3.65V is right around the 40% SOC mark, I believe and 3.80V is right around 50% (I need to use ScanMyTesla to confirm). Tesla, drop the minimum charge level to 40%!

Radware Bot Manager Captcha

Dr. Jeff Dahn consults for Tesla and is one of the foremost authorities in advanced battery research.

Tesla gets a new head of its Advanced Battery Research in Canada, renews contract

Tesla now has a new head with its Advanced Battery Research group in Canada and renewed the contract with the organization, which has been producing a lot of battery technology for the automaker. Tesla Advanced Battery Research For the past five years, Tesla has been working with a battery...

electrek.co

"For the past five years, Tesla has been working with a battery research group in Canada, also known as Jeff Dahn’s research team, that has been helping improve their battery tech. Dahn is considered a pioneer in Li-ion battery cells. He has been working on the Li-ion batteries pretty much since they were invented. He is credited for helping to increase the life cycle of the cells, which helped their commercialization. His work now focuses mainly on a potential increase in energy density and durability, while also decreasing the cost.

In 2016, Dahn transitioned his research group, which operates as part of Dalhousie University in Halifax, Nova Scotia, from their 20-year research agreement with 3M to a new association with Tesla under the newly formed “NSERC/Tesla Canada Industrial Research.” The partnership has already been very beneficial for Tesla. Over the last few years, the group has been filing patents on battery technology for Tesla that could lead to a new battery cell enabling a million miles in a battery pack."

 

[Steve446]

Very interesting study from Jeff Dahn et al on how limiting cell voltage on common NMC chemistries to 3.65V will lead to vastly improved lifetime of batteries. and a NMC battery designed for 3.8V max operation beats LFP in energy density and is projected to last nearly a century at room temperature.

Note - the large increase in battery life also should apply to any NCO/NMC cell charged to a lower maximum voltage. 3.65V is right around the 40% SOC mark, I believe and 3.80V is right around 50% (I need to use ScanMyTesla to confirm).

Tesla, drop the minimum charge level to 40%!

Radware Bot Manager Captcha

I confirm those V vs SoC for my car from my SMT data although the curve is very straight around those voltage values (I think @AAKEE also showed a publication with a graph like that?)

 

[AAKEE]

I confirm those V vs SoC for my car from my SMT data although the curve is very straight around those voltage values (I think @AAKEE also showed a publication with a graph like that?)

We need to define how we measure the voltage or know that we use the same or different reference.

Theres only one solid defined voltage for a specific SOC, thats the OCV. With no load on the battery the voltage creeps up to the defined value after a period of rest.

Driving the car lowers the voltage depending on the power output. In Park after a drive the voltage is also lower due to the not yet recovered voltage and the load from the car in park.

At all loads the cell voltage bogs down and the bog varies with cell temp, load history etc.

This is a OVC voltage chart for a NCA cell (Panasonic 18650, I think).

Tesla on screen SOC is slightly lower than true SOC, due to the buffer below 0%.

TRUE SOC: (4.5 - screen SOC x 0.045)+ screen SOC

I have a lot of screen dumps from SMT.
A couple of weeks since new:
3.82 average cell voltage when the car had been standing parked in the garage for about 5-6 hours, 54.72% SOC.
Not OVC, because the contactors close when unlocking the car.

After my regular 55% charges (thats about 57% true SOC) I can see the cell voltage 3.84v after the voltage drops from the charging session when using the log data of teslalogger.

 

[ditchla]

Really how much truth is there to this battery charging nonsense? Don't charge to 100%, dont drain below a certain percent.

What do you think is going to happen to my batteries? Really, anyone have experience with this? I will share my side below, since I think a lot of people out there are scared to even try this for a week. I've done it for a year and driven A LOT.

I have had my Model 3 performance for about 1 year now. Its got 24k on the odometer. Obviously a daily driver.
My charging habits.
Charge to 100% every night (home charger)
Drive it down to whatever it takes to finish the day. Sometimes 10 miles left sometimes 20 miles left sometimes 80 miles left, then back home to the charge to 100% for the next day. Again, EVERY night at my house it plugs in and charges to 100%. I have the $500 tesla wall charger.

When I bought the car I needed those miles because of all my driving. I literally only bought the performance for the extra miles, I would have gladly taken the long range but the wait was too long.

I never realized the games Tesla wants us to play. Only charge to 80% and only run down to 40% or 20% or blah blah blah. If I knew that I would say the car is a bait and switch. So what did I decide to do? Just drive the car and use the car. Like a gas car I fill it up run it to E like a gas car, and like a gas car I've hit the E in my Tesla and driven to the "gas" station with nothing in the tank.

I've brought my tesla down to 0 miles 2 times since I owned it. THIS I avoid if I can because I cant just get some gas and drop it into the tank, I would have to find a plug and sit there for hours. Last time it happened was scary but let me offer up this info to the forum. At 0 MILES I could still drive 4 miles (3.5 at 60 mph on the freeway and .5 to the charger) Again. SCARY on the freeway at 0 miles
However it would be nice to know WHAT I could actually get at zero miles, on my gas cars at E I usually knew 5-10 miles was easy peasy.

Also you might ask. WHY? Why drive this way and not follow the rules by Tesla and some owners advice online....Why? Cause I dont have time to sit back and watch Netflix and chill at chargers around town 2 times a day. I would rather I charge at home. I would rather "destroy" the batteries then spend 20-30 min a day, almost every day, at a charger as to not dip below a certain percentage.

Also, I keep cars for 2-3 years. MAYBE 4. Is any of the above going to matter in the short term (3-4 years)? Will I have to replace the batteries? I dont know. I just dont. There is so much rhetoric and battery virtue signaling its hard to tell whats real. Hence this post.

What I CAN tell you is the car works the same as the day I bought it, charges the same, and same basic range since the day I bought it. So no noticeable battery issues from my habits....so far 1 year in and 24k.

 

[Bouba]

I’m not an expert but I follow the experts on here religiously....the death to batteries is storing the car with a full battery...your charging habits are charging then using so you may be alright