navguy12
Active Member
My latest slow, overnight trickle charge before departure on a long trip (picture from two days ago):
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health
- Thread starter KK_RedM3
- Start date
My Model 3 RWD shows that charge stats in the app as "Total Charged" = 1400 kWh.
While my car running stats show that I have utilized 1125 kWh.
My car has run for 4400 miles till date.
Does it mean that the car is using ~20% in pre-heating/sentry-mode and other idle conditions?
And Assuming a battery of 57.5 kWh,
Is my car battery is giving average run of 4400/1400*57.5 = 180 miles
instead 4400/1125*57.5 = 225 miles
As against what they claim as 267 miles under standard conditions.
While my car running stats show that I have utilized 1125 kWh.
My car has run for 4400 miles till date.
Does it mean that the car is using ~20% in pre-heating/sentry-mode and other idle conditions?
And Assuming a battery of 57.5 kWh,
Is my car battery is giving average run of 4400/1400*57.5 = 180 miles
instead 4400/1125*57.5 = 225 miles
As against what they claim as 267 miles under standard conditions.
My Model 3 RWD shows that charge stats in the app as "Total Charged" = 1400 kWh.
While my car running stats show that I have utilized 1125 kWh.
My car has run for 4400 miles till date.
Does it mean that the car is using ~20% in pre-heating/sentry-mode and other idle conditions?
And Assuming a battery of 57.5 kWh,
Is my car battery is giving average run of 4400/1400*57.5 = 180 miles
instead 4400/1125*57.5 = 225 miles
As against what they claim as 267 miles under standard conditions.
Yeah, Sentry, cabin overheat protection, preconditioning, and natural battery losses from the car just sitting aren’t factored in to EPA numbers, so it can be hard to hit EPA range without disabling those features or doing all of the driving in one go or over a short span of time. My RWD has only used 72% of total energy added to the battery for driving, but that doesn’t surprise me because I work from home and sometime take several weeks or more to go through an entire battery worth of energy, so I lose a lot to natural battery losses over time, and I also use Sentry a lot. My numbers are 9801 miles / 3026.2 kWh total energy added * 57.5 kWh usable battery capacity = 186 miles, or 9801 / 2185.6 * 57.5 = 258 miles if you only consider energy used while driving. Sounds about right. I have actually driven 259 miles on a single charge, and that included some idling over a couple days, and I still had a few percent left when I got home, so I know the car can do the EPA rated range in the real world if you do all of your driving in a relatively short span and keep the speed under 70 mph.
Last edited:
AlanSubie4Life
Efficiency Obsessed Member
I don't know that the data supports 90-100% for brief periods being significantly worse than 70%-90%.
Seems like just charge to 100% whenever you need it and draw it down rapidly to below 55% for NCA.
And keep cycle depth minimum of course.
If you have to do deep cycles to use the car it seems silly to limit your utility. Of course if you don't need 100%, don't do it, but having margin is important sometimes.
It's discussed well elsewhere in this site. For aftermarket, get Hankook iON EVO All Season tires for superb range/efficiency and good noise absorption.
Dr Jeff seems to only talk stress and cycles, and seldom cover* anything else but the cyclic aging.
Calendar aging causes a lot of degradation during the first 5-8 years of the battery life.
In fact, cycling aging only causes a fraction of that for the most owners.
Panasonic ncr18650 after 18 months: 100% is not the worst SOC.
Actual model S cells:
Charts showing cyclic aging shows that the degradation per cycle converted to miles causes about 0.5% cyclic aging per year.
—> If anything, if any owner would like to minimize degradation he/she should take the actions that reduce the calendar aging.
Doing this will also keep the cyclic aging low.
We know from research/tests that Panasonic NCA degrade less from calendar aging at or below 55% displayed on a Tesla.
The cyclic aging is also smallest between ~55 to ~30% displayed SOC so it matches good.
Actual model 3 cells cycled:
The SOC shown is true SOC, and because of Teslas buffer, the 5-15% is 0-10% displayed.
It might look bad for 5-15%, but actually those cycles 17% loss after 3000FCE is = 17% after 750K miles, so thats atound 60 years of average driving or 0.28% per year of cyclic aging.
Calendar aging is less the lower the SOC is, down to 0% so Dr Jeff might use a ”medium” SOC as the ideal, taking in margins to ensure not to find the car empty when arriving at home.
IRL, the HV-lithium battery will degrade less the lower the SOC is, so 30% is not the sweet spot for low calendar aging - 30% might be a compromise also making Dr Jeff sure no one will sue him one day for the statements.
There is no bad thing for cyclic aging, going low in SOC. In the fine scales charts we see slightly more degradation from cycling below 20% displayed SOC but the increased wear is lower than the reduction in calendar aging, so still a win to go low.
More or less any research shows that calendar aging is doubled above 55% displayed SOC so a recommendation to charge to 65-75% will not be the best thing for degradation.
Keeping the battery at or below 55% as often as possible will be the best solution.
Things to know to do extra well with the tips above:
- Going down to 0% is safe for the HV battery. When needed, have a margin to surpress your own range anxiety.
- Charging to 100% is not that bad for the battery. You do not need to drive asap, at least no more need than when charging to 80%.
*) I recommend strongly to widen the sources for knowledge about lithium batteries.
Dr Jeff might be good, but there are several others of the same dignity that have produced research/reports.
A single source might lead to a skewed picture of this.
The problem I had with Dr Jeff is that he gave information about which cells perform better or worse...but was very vague on how to find out which cells are in your car....I realize this information is not given freely by Tesla....but do you have a source which can tell which battery has NMC532 or NMC622 ?
All Teslas historically used Panasonic NCA. Thats not NMC.
All S/X still use NCA.
Model 3/Y was NCA untill ~2021/2022 in rest of the world but continued with NCA in USA.
The LG 78.8 is now used in USA as well for model 3 LR and as it seems also in the Y LR.
The LG 78.8 (5L in European reg.) is NMC 811 ( but actually about 80% Nickel, 8% Manganese and 8% Cobalt and 4% Aluminum).
I have seen Dr Jeff talk about “wrong” chemistry but as I see him as only one of many researchers he do not get more space in my book than other researchers/reports.
I build my own picture and Dr Jeff’s presentation do not always match what other research reports find. It might be more a thing about what to focus on (just as I wrote earlier).
So, this is actual test data for LG Chem 21700 M50 cells. Thats more or less the same as the E5LD/Y5LD in European terms, LG 78.8kWh pack that we now also see in US on 3LR ( and as I take it, also on Y LR).
Can not guarantee that it is exactly the same as innthe cars, but it should be closely enough related to tell us what we need to know about the aging of these cells.
The colors chosen are little too close to easy distinguish it, but after looking a bit more in detail we can see that this battery degrades from calendar aging precisely as expected ( from older tests of other NMC cells.
-Low SOC gives very low calendar aging, around half degradation from the high SOC region.
- The step with the change from low to high calendar aging is as expected where 60% has Low degradation and 70% are beyond the step and has high degradation.
-Once again, 100% is not the worst. 70-95% seems to have about the same degradation and 100% slightly less than that group (25C aging).
The resolution of this test is 10% Sspc steps around 60%, so we get no proof of the central graphote peak tomve at 62%, on the other hand we get no indication that is does not. As 60% even seems slightly better than 50% we can estimate that 62% is OK also.
So, for the LG pack we can safely say that 58% displayed SOC or below is good and that most probably, 60% or below should be good.
(The reason for this post is that I used older research to make a qualified guess that the LG 78.8kWh pack would be at low calendar aging at 60% displayed SOC, but that was actually a qualified guess. Now we have data that both supports that and proofs that this actually is the case.)
(Source: Lg M50 research )
Can not guarantee that it is exactly the same as innthe cars, but it should be closely enough related to tell us what we need to know about the aging of these cells.
The colors chosen are little too close to easy distinguish it, but after looking a bit more in detail we can see that this battery degrades from calendar aging precisely as expected ( from older tests of other NMC cells.
-Low SOC gives very low calendar aging, around half degradation from the high SOC region.
- The step with the change from low to high calendar aging is as expected where 60% has Low degradation and 70% are beyond the step and has high degradation.
-Once again, 100% is not the worst. 70-95% seems to have about the same degradation and 100% slightly less than that group (25C aging).
The resolution of this test is 10% Sspc steps around 60%, so we get no proof of the central graphote peak tomve at 62%, on the other hand we get no indication that is does not. As 60% even seems slightly better than 50% we can estimate that 62% is OK also.
So, for the LG pack we can safely say that 58% displayed SOC or below is good and that most probably, 60% or below should be good.
(The reason for this post is that I used older research to make a qualified guess that the LG 78.8kWh pack would be at low calendar aging at 60% displayed SOC, but that was actually a qualified guess. Now we have data that both supports that and proofs that this actually is the case.)
(Source: Lg M50 research )
So, for the EU Model 3/Y they use NMC811 (with a thad of aluminum).
Se post above.
That research report mainly research charging approaches connected to V2G but the real calendar aging and cyclic aging results tell us quite much:
Despite that they only tested the cycles as 100% to 0% SOC cycles; they found that the cyclic aging was very small compared tonthr calendar aging for normal use of the battery.
So, the bottom part shows us the calendar and cyclic aging for a ”gentle driving profile”.
What we can see is that the cyclic aging is more or less negligible compared to the calendar aging.
In this case, would we like to focus on the cycles or would cutting the calendar aging in half be a better idea?
Quite obvious, I guess…(to add on, the way we reduce calendar aging also reduce cyclic aging).
Back to the Dr Jeff statements: Listen and read those, but do not focus only on one single researcher - you might find yourself trying to reduce microcracks and might fail to see the big picture.
AlanSubie4Life
Efficiency Obsessed Member
Unlikely unless sourcing has changed. 2024 Model Y LR qualifies for POS tax credit.
So battery must meet US gov’t sourcing requirements. Which LG packs have not met at least not to our knowledge.
Packs are frequently coming in at >=79.5kWh, 304 miles, (4 out of 4 vehicles checked when my dad took delivery of his 2024 Model Y LR).
Would be very interresting to se SMT hooked up to one of these cars.
oreobbq
Member
Good to know. This is in line with everything else I’ve seen about these batteries. Heck my iPhone caps the charging at 80% to prolong battery health.
Maybe he’s the foremost battery expert but only quack doctors refer to themselves using their first name. Dr. Phil etc. Haha
This is now my favorite way to spell this word.
Possibly the Y LR packs are still LG NMC but made in Korea or USA which might qualify for the incentive?
Most likely all Model Ys, RWD/LR/Performance, that are made in Fremont, and Texas, for sale in the US, use Panasonic cells made at Giga Nevada.
Then why don't they have 82 kWh like the other Panasonic NCA packs? That's the question. Particularly with the less efficient Y they would want to have a higher energy if they could.
Similar threads
