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

Hearing super conflicting advice about battery charging levels

This site may earn commission on affiliate links.
Hi,

I’m a new TM3 LR AWD owner. I’ve been browsing these forums regarding what % to charge to. I can’t find a consensus. I see some people in the forum advocate to charge to 80-90%. Some advocate to maintain an average of 50%. And I’ve even seen some say the lowest SoC is best.

I guess what I’m most interested in is how much of a difference does it actually make? Like if charging daily 50% means I get 5 more miles compared to to the 80% guy after X years then I would rather just charge to 80%.

I plan on owning this car for a long time so I want to make sure I’m not damaging it. I’ve been charging to 80% daily but now I’m not even sure if that was a good idea since I use like 10%-15% a day.
 
Higher states of charge induce voltage stress to Lithium-ion battery cells. A 100% SOC equates to ~4.2v/cell and provokes significant stress. Reduction of voltage/cell proceeds in a roughly linear fashion equating to lower SOC's, with steadily improving life cycle counts in NCA chemistries. At ~3.92v/cell little further improvement in life cycle counts are seen.

3.92v/cell equates to ~60-65% SOC. So using 65% SOC as a charge ceiling essentially eliminates voltage stress as a significant factor in battery degradation.

For those who routinely charge above ~65%, one can assume some amount of battery degradation due to voltage stress. That can be observed as a fairly linear reduction in battery lifecycle counts as the percentage ticks higher. How much that matters - or if it matters at all - is up to each owner to decide.

Bear in mind that there are other factors beyond charge level that also affect pack longevity. So it becomes a somewhat more complicated picture.
 
Hi,

I’m a new TM3 LR AWD owner. I’ve been browsing these forums regarding what % to charge to. I can’t find a consensus. I see some people in the forum advocate to charge to 80-90%. Some advocate to maintain an average of 50%. And I’ve even seen some say the lowest SoC is best.

I guess what I’m most interested in is how much of a difference does it actually make? Like if charging daily 50% means I get 5 more miles compared to to the 80% guy after X years then I would rather just charge to 80%.

I plan on owning this car for a long time so I want to make sure I’m not damaging it. I’ve been charging to 80% daily but now I’m not even sure if that was a good idea since I use like 10%-15% a day.

The reason there "is no consensus" is because there isnt one. This entire thread is about basically the same general discussion:


The "BEST" thing is to set the charge level at 50%, leave it plugged in, and never ever drive it. Failing that (because its silly suggestion), if you want more factual information than you could ever want on this topic, a new thread is not needed.

Simply search the above thread for posts from @AAKEE , and @AlanSubie4Life . If you do so, you will get more information than you likely could possibly want on this topic that is factual, not tinged with superstition or bias, and will help you make a decision on your own on what to do.

There is nothing that anyone could post in this new thread, that has not already been said in the above one in general, and the posts in that thread from those two individuals have a TON of responses to this EXACT question, posed by a ton of people, because I used to put all threads like this created by people into that one (to try to get people to read the already posted answers by them).
 
There's no consensus because there's no one single answer that is best for all aspects.

Battery longevity is only one aspect you can optimize. There are research papers on that subject and you can figure out from them that the ideal SOC is around 50%, always doing small charges, and keeping the battery around 20C. You'll also see that supercharging might be harder on the battery than AC charging at home.

Then there's usability of the car. You might have an emergency that requires that you drive far, at a moment's notice. Keeping the battery at 50% might not be best. The user manual will recommend something like 90% SOC with daily charges at least partly because of this.

Then there's battery capacity estimation. The battery management system (BMS) can only provide a good estimate of total capacity and current state of charge if it regularly gets good readings. It needs to see low voltages (low SOCs) and high voltages (high SOCs), while the car is sleeping for a few hours. To optimize this, you leave the car at 10% for a night sometimes, you leave it to 90+% for a night other times etc. This is conflicting with the longevity advice.

The new LFP battery that's in the 2022 RWD model 3s is even worse as the voltage difference between a low and high soc is smaller. That's why the manual recommends charging to 100% weekly (just for LFP). That's not best for the battery longevity but it's best for the BMS which otherwise might estimate so badly that you see 5%SOC and are suddenly stranded, out of juice.

The battery longevity research papers will help you figure out what the actual difference is. I have seen someone put that in kilometers and it was something like 3% capacity difference after 200k kms at 25C temperature. Higher temps will make a bigger difference.

If all this interests you, go read the multiple posts on the subjet in the model 3 battery forum. If not, don't worry too much about it.
 
With an NCA battery (as the LR model has), high states of charge increase the rate of capacity loss, so it is likely that the best daily charging method is to set the charge target to B + D, where B is your preferred buffer against running out and D is the amount you use during the day (including the effects of climate control use, weather, sentry / cabin overheat / dog / etc. modes, and side trips). For example, if B = 10% and D = 45%, set the charge target to 55%. Obviously, if this is less than 50%, use 50% because that is the lowest allowed setting. In addition, set it to finish charging just before you leave for the day to minimize time parked at high state of charge (this is more important if your charge target is higher).

Obviously, charge it higher for longer trips, and it may be necessary occasionally to charge from low to 100% to ensure that the BMS has not lost the true state of charge. LFP batteries (as the 2022 RWD model has) may need the latter more frequently (Tesla says weekly), but are likely to lose capacity at a slower rate than NCA batteries, so 100% state of charge is likely less detrimental to them than for NCA batteries. If you follow the above for an LFP battery, it may be prudent to use a higher B to avoid running out because the BMS became inaccurate.

Now, how much that matters in actual cars (versus in battery labs) compared to other charging methods is something that may be harder to figure out, since other peoples may have other differences besides charging methods.
 
  • Helpful
Reactions: Transformer
For your specific situation: If you use only 10-15% each day, you can charge to 50% each day and come home with 35-40% at the end of the day, so that seems like plenty of safety buffer against running out (unless you may do unplanned longer drives during the day). Seems like little or no effort to do so, other than resetting the charge target when you need more charge for a longer drive the next day.
 
There's no consensus because there's no one single answer that is best for all aspects.

Battery longevity is only one aspect you can optimize. There are research papers on that subject and you can figure out from them that the ideal SOC is around 50%, always doing small charges, and keeping the battery around 20C.

Thats not correct.

All research show that the lower the SOC the less is the degradation. Down to 0%.

This is a fact (as we can consider research and science that is very consistent = facts) and not a forum truth.

As for the temperature, during rest = lower cell temp will reduce degradation.
Research show that at around -20C the degradation is virtually stopped.

For cycling, the optimum temperature is somewhere around 25-30C, in some cases even higher.
For supercharhing 50C causes much less wear than 20-30C.

One of the ”big problems” we have about batteries is that there have been forum truths that probably occured because people did read teslas advice and than added some own ”facts” to this which in fact in many cases is not correct.

One example is that (for NCA/NMC Teslas) daily charge is to 90% and 90-100% is preserved for traveling/longer drives.

The forum says that you better not leave the car more then very very short time at 100% because otherwise your battery take a big hit.
Even though this is a good idea to follow, tesla do not say this in the manuals.
Actually, the research do not show that 100% is much worse than 80% and actually some research about 2170 NCA cells show that the worst degradation happens around 80%.

Also, the forum usually say that going below 20% is bad for the battery.
Tesla never said this in the manuals. For leaving the car at low SOC for longer time they refer to 0% and use 1% per day away as a way to calculate the minimum SOC to leave the car. Tesla refer to other components that can be damaged if the main battery disconnects, as the low volt battery(lead acid) that really takes a hit from going below 50% SOC.
Still many people is afraid going below 20% when the real world actually is that low SOC causes less wear.
 
For your specific situation: If you use only 10-15% each day, you can charge to 50% each day and come home with 35-40% at the end of the day, so that seems like plenty of safety buffer against running out (unless you may do unplanned longer drives during the day). Seems like little or no effort to do so, other than resetting the charge target when you need more charge for a longer drive the next day.

I did have about this daily charge schedule until about now (changed job).

Charged to 55% (to stay below the central graphite peak, search for my nick and that term if info is needed) and had 60miles/95km daily drives back abd forth the job.
Depending on summer (+20C) or winter (-30C) i arrived with 20-35% home. Most days I did some additional drives during the evening, to friends, the supermarket etc.
The car often slept the night at about 25-30%.
( I could have set 50% but from all I learned about the calendar aging I decided to set it to 55% which gives around 57% true SOC and by this still stay below the step that increases the degradation. This gave me more freedom of action for a very low increase in degradation)

Did not charge until the charge(WC 11kW) had to start to be commenced before the next morning so it started charging about 0300-0430 in the morning, set via planned charging.

My average SOC was somewhere around 35% during a quite long time.
Average battery temp over almost one year was 10C.

After 42000km I still had full range(507km/315 mi M3P 2021).
In the real world we should know ( I know) that calendar aging will eat a cake from the battery even if the BMS can not find this loss.
I did some reversed BMS Calibration and managed to get the BMS on track(as I think) at 79kWh. My calculations said that that the capacity should be around 79kWh and the BMS stopped at 79.2kWh.

The degradation is lowest of the M3P ’21s in teslafi.

My new job is 240km single way but a one week stop on job before going home.

At home I still use 55%, but the average SOC is higher as I do not drive much.
Average SOC for the latest month was 47%.
 
I did have about this daily charge schedule until about now (changed job).

Charged to 55% (to stay below the central graphite peak, search for my nick and that term if info is needed) and had 60miles/95km daily drives back abd forth the job.
Depending on summer (+20C) or winter (-30C) i arrived with 20-35% home. Most days I did some additional drives during the evening, to friends, the supermarket etc.
The car often slept the night at about 25-30%.
( I could have set 50% but from all I learned about the calendar aging I decided to set it to 55% which gives around 57% true SOC and by this still stay below the step that increases the degradation. This gave me more freedom of action for a very low increase in degradation)

Did not charge until the charge(WC 11kW) had to start to be commenced before the next morning so it started charging about 0300-0430 in the morning, set via planned charging.

My average SOC was somewhere around 35% during a quite long time.
Average battery temp over almost one year was 10C.

After 42000km I still had full range(507km/315 mi M3P 2021).
In the real world we should know ( I know) that calendar aging will eat a cake from the battery even if the BMS can not find this loss.
I did some reversed BMS Calibration and managed to get the BMS on track(as I think) at 79kWh. My calculations said that that the capacity should be around 79kWh and the BMS stopped at 79.2kWh.

The degradation is lowest of the M3P ’21s in teslafi.

My new job is 240km single way but a one week stop on job before going home.

At home I still use 55%, but the average SOC is higher as I do not drive much.
Average SOC for the latest month was 47%.
Gotcha. I will set it to 50-65% for weekdays and higher like 80 or 90% for weekends. That’s probably a good balance between real world use and minimizing battery loss
 
Higher states of charge induce voltage stress to Lithium-ion battery cells. A 100% SOC equates to ~4.2v/cell and provokes significant stress. Reduction of voltage/cell proceeds in a roughly linear fashion equating to lower SOC's, with steadily improving life cycle counts in NCA chemistries. At ~3.92v/cell little further improvement in life cycle counts are seen.

3.92v/cell equates to ~60-65% SOC. So using 65% SOC as a charge ceiling essentially eliminates voltage stress as a significant factor in battery degradation.

For those who routinely charge above ~65%, one can assume some amount of battery degradation due to voltage stress. That can be observed as a fairly linear reduction in battery lifecycle counts as the percentage ticks higher. How much that matters - or if it matters at all - is up to each owner to decide.

Bear in mind that there are other factors beyond charge level that also affect pack longevity. So it becomes a somewhat more complicated picture.
Oh gotcha. Sounds like 65% is a sweet spot to use then! Maybe I can do 65 on weekdays and 90 on weekends
 
Oh gotcha. Sounds like 65% is a sweet spot to use then! Maybe I can do 65 on weekdays and 90 on weekends

This is how calendar aging losses looks, SOC vs time and temperature:
6DD6C4E4-54E0-4922-A641-E4B0F0B120EA.jpeg


LR and P use NCA. ( China and Europa built get NMC)

There is a clear benifit to stay at/or below 55%. The ’step’ that increases degradation from time is confirmed bu numerous reserch tests. (There is also research reports masking the step by the test setup, like only test at 20, 50 and 90% SOC and wrongly draw a straight line between these points).
The step occurs due to the ”central graphite peak” which van be googled togheter with ”NCA” and ”calendar aging”.

65% is motivated if you need the range, but it will cause almoste the double degradation compared to 50-55%.

Remember that calendar aging SOC numbers occurs on the formula [TIME x TEMP x SOC].
If the time is nulled out after a 100% charge by driving away immediately there is virtually no calendar aging at 100%, and only the cyclic wear counts.
That said, the calendar aging at 100% isnt much worse than 70-90%, in fact there is research that show 80% as the worst SOC.

My car get time at 55% during wekeends. I like to have the freedom of range (55% is enough for me). Its charged saturday (and sundsy) morning to 55%, and then various use. During weeks it sleeps at 20-35%, and get charged just in time before the drive.
 
This is how calendar aging losses looks, SOC vs time and temperature:
View attachment 829107

LR and P use NCA. ( China and Europa built get NMC)

There is a clear benifit to stay at/or below 55%. The ’step’ that increases degradation from time is confirmed bu numerous reserch tests. (There is also research reports masking the step by the test setup, like only test at 20, 50 and 90% SOC and wrongly draw a straight line between these points).
The step occurs due to the ”central graphite peak” which van be googled togheter with ”NCA” and ”calendar aging”.

65% is motivated if you need the range, but it will cause almoste the double degradation compared to 50-55%.

Remember that calendar aging SOC numbers occurs on the formula [TIME x TEMP x SOC].
If the time is nulled out after a 100% charge by driving away immediately there is virtually no calendar aging at 100%, and only the cyclic wear counts.
That said, the calendar aging at 100% isnt much worse than 70-90%, in fact there is research that show 80% as the worst SOC.

My car get time at 55% during wekeends. I like to have the freedom of range (55% is enough for me). Its charged saturday (and sundsy) morning to 55%, and then various use. During weeks it sleeps at 20-35%, and get charged just in time before the drive.
Thank you! I’ll set it to 55% then. I assume even if I average more than 20% a day, it’s not bad if the mean battery percentage is below 50% right?

I guess the only other question is how much does 55% SoC affect accceleration since that does give me enjoyment
 
Thank you! I’ll set it to 55% then. I assume even if I average more than 20% a day, it’s not bad if the mean battery percentage is below 50% right?

I guess the only other question is how much does 55% SoC affect accceleration since that does give me enjoyment
It's not bad that the battery is that low, but acceleration is going to suck. I've noticed a significant hit to performance under ~60%. Under ~20% is like being stuck on chill mode.
 
  • Like
Reactions: Rodney!
Thank you! I’ll set it to 55% then. I assume even if I average more than 20% a day, it’s not bad if the mean battery percentage is below 50% right?

I guess the only other question is how much does 55% SoC affect accceleration since that does give me enjoyment
Anything on the low side of 55% is good.
Even down to 0%.

The loss of power depends on what car, and the preferences.

I have a 21 M3P and since they tuned the software (probably after having enough data on the 2170L) my car performs wuite ok down to 50% or lower.
The battery temp if low can limit the power.

I think I did 3.60 0-100kph at about 50% SOC with the dragy app/box.

A regular LR should still deliver full power (but is 1/2 to begin with, so ;) )
 
All research show that the lower the SOC the less is the degradation. Down to 0%.
Whoa, I've seen this before. I thought there was a dip in the low-middle, where least degradation was somewhere in the approximately 30%-40% level or so. Going down near 0% for storage should not be least degradation, as it really does have other kinds of damage being down that low.

This is how calendar aging losses looks, SOC vs time and temperature:
I tried to look at those, but your picture only shows the top row of graphs and has the bottom row of graphs cut off in the picture so we can't see them. The top row is about temperature, but the bottom row (the ones I want to see) are about state of charge storage and aren't showing all the way to read them.
 
Whoa, I've seen this before. I thought there was a dip in the low-middle, where least degradation was somewhere in the approximately 30%-40% level or so. Going down near 0% for storage should not be least degradation, as it really does have other kinds of damage being down that low.


I tried to look at those, but your picture only shows the top row of graphs and has the bottom row of graphs cut off in the picture so we can't see them. The top row is about temperature, but the bottom row (the ones I want to see) are about state of charge storage and aren't showing all the way to read them.
Its all there:

This is the important part for NCA cells cut out.

060FABF4-F3AA-4919-92A5-4B776CB79395.jpeg


For starters:
SOC is defined by cell voltage.
100% SOC is the maximum alloved cell voltage. (4.20V for NCA)
0% is the minimum allowed cell voltage, or the minimum discharge voltage. 2.5V for NCA.
This means= 0% is not empty, but it is the end of discharge point according to the cell manufacturer.

0% is safe. Period. There have been a lot of research tests. Down to 0% does not cause any damage, in fact cycling 10-0% most often made the batteries do the most cycles.
Going below 0% is overdischarge and this causes random damage to the cells. Some cells breaks, some cells doesnt seem to care and come out with zero loss and some get heavy degradation. Overdischarge is not good.

The graph show the loss during 10 months for three different temperatures. The cells is put at the test SOC for a couple of months, then cycled a few cycles to ”wake up” and then tested, after this put to the same SOC and then stored a couple of moths: repeat.

”I dont store my car for 10 months”

- No, you do not. The average driver do 10K miles or so in a year. This at maybe 30 miles avg speed. This set us at 333 hours of driving each year, or about one hour a day. Daily drives = 27 miles = 1-2hpurs of charging each day. Lets say 1.5.

Now the sum is 1.5x365 + 333= 880 hours of ”cycling”.

One year is 8760 hours, so there is 7880 hours when the car sleeps. This is equivalent to the storage in the test.
This is also 90% of the time and > 10 months.

Calendar aging reduces with time. The first year is the worst and each year on take a progressively smaller bite of the battery.
( reduction about square root of time )
 
  • Informative
Reactions: Steve446
Overdischarging (discharging below the manufacturer low voltage limit:

The 2.7V is the normal end of discharge fpr this LCO battery.
Overdischarging to 1.5V (thats quite massive, and not possible in a Tesla*) just caused a minor drop in the number of cycles from cirka 160 to cirka 145.

Overdischarge to 0.0V is clearly not good.

197EE530-A8E8-408F-A187-5DDED7993FB2.jpeg


ShieldSquare Captcha

0% in the Tesla screen SOC is about 4.5% true SOC as the buffer is below zero.

Driving down to 0% means true SOC is about 4.5%. Not dangerous. Safe for the battery.

The BMS will shut down the battery to protect it when the SOC /cell voltage has reached the low voltage limit.
After that the cells are at about 0% SOC or slightly above for safety, and the self discharge is low.
 
There is a ton of research.

Evelina Wikner has done a lot of research and the reports are easy to read.

https://research.chalmers.se/publication/512004/file/512004_Fulltext.pdf

There can really be no doubts after seeing data like this in a lot of reports (but better presented by Evelina)

0-10% cycles can not be ”damaging the battery” if the battery holds up 20C better than 60-70% cycles and so on.

One FCE is the energy of 100-0% in the battery.
This means that 10.000FCE actually is 100.000 0-10% cycles!!!
(Chart use FCE)
E1685C95-27DF-4EB5-BDC3-E1DEDCBF4668.jpeg