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

[Xebec]

Dave's recent post reminded me that I wanted to ask about this graphic that AAKEE posted. It seems to show a sharp break between aging about 55% and lower, versus aging at 60% and above. The text to the right of the graphic seems to address this a little, but, it's hard to read since it's cut off.
View attachment 721862
Can anyone explain what is happening in this region?

I’ve been thinking about this and AAKEE’s posts a bit more. AAKEE pointed to a really good study here:

https://publications.lib.chalmers.se/records/fulltext/249356/249356.pdf

If you go down to section 5.2.1 (page 50), they have a similar chart/graph but out to 700 days or about 2 years. They tested calendar aging at 15%, 75%, and 90% SOC, and at 25C, 35C, and 45C.

Reading the graph, I’m concluding two things: 1. At 25C there was no appreciable difference in aging between 75% and 90% SoC. I.e. If you’re in a moderate climate, then the difference in battery calendar aging of 75% to 90% SoC is minimal to zero, maybe not worth dropping the charge down.

2. At 2 years in, the 15% SoC appeared to be at 98.5%-99% capacity, while 75% (and 90%) @ 25C appears to be 94.5%-95.0% capacity. Perhaps the inflection point is 55% as per the study above. If that’s linear over time - then at 10 years you’re looking at 5%-7.5% loss staying at 15% SoC, and maybe 25-27.5% loss at 75% or 90% SoC. (At 5 years the losses would be about 3% and 13% respectively).

It feels like if you’re going to keep the car for 5-6 years, I don’t see “3% degradation vs 13%” making much of a difference in resale value. For road trips at 5 years old, the 10% difference probably is tolerable too.

However, if you are keeping the car a lot longer - 10 years or more; at 10 years you probably also don’t havre too much difference in resale value - because 10 year newer EVs are going to be much better (and cheaper) in the battery department.. It’s going to be hard to trust that battery degradation really is low on a 10 year old vehicle even in the face of evidence, and “feeling” about a car tends to determine it’s price at that age. OTOH a road trip in a 10 year old EV with 25-30% range loss is huge vs. one with 5-7% range loss.

(I normally keep cars for 12-15 years myself (until they’re basically costing a lot of money and time to maintain), but with EVs improving so much every year especially over the next decade, I might revise that. Sell the 4-5 year old EV while it’s still worth something or similar. )

All this said I’m still thinking about limiting to 55% SoC based on the data above going forward to prevent further degradation of my 3…

 

[AAKEE]

Interesting. So if I have 100k miles on my M3 and only have 77 "usable" kWh now (for example) due to degradation, you're saying there may not be much "left in the tank" so to speak after 0%. Whereas someone with a new M3 probably has a good bit left after 0% SOC.

Nope.

The 4.5% Buffer is always 4.5% of the (by BMS) estimated capacity.
But the thing that can happen is that the difference in cell size and imbalance between cells cause the voltage to drop in some of the cells , and when the lowest cell reach the minimum voltage this causes the battery to shut off.

If you recently did a full charge which include cell balancing, the balance at the top( 100% SOC) is good, causing the balance in the bottom to get less good.
This invrease the probability that the car shuts down early.

One should newer calculate to be able to use the buffer. 20mV imbalance with low SOC is probably not unlikely, and this will count for about up to 2% of the capacity = only about 2% of the buffer will be available.

 

[AtticusRex]

What should be the range on a Model 3 Long Range with AWD? It has 14K miles and was purchased in November of 2019.

I’ve just decided to charge my Model 3 to 100% every time. It’s now down to 263 on 90%.

 

[Jeremy3292]

I’ve just decided to charge my Model 3 to 100% every time. It’s now down to 263 on 90%.

This is not a good decision for longevity of the battery. Do you need 100% every day? What's wrong with 90% or 80%?

 

[AtticusRex]

This is not a good decision for longevity of the battery. Do you need 100% every day? What's wrong with 90% or 80%?

What am I worried about? Losing battery capacity? It already loses 1% a month.

 

[Jeremy3292]

It already loses 1% a month.

Probably bc you charge to 100% every day which Tesla specifically says not to do unless you need it for long trips.

You can certainly do whatever you want as it is your car.

 

[AtticusRex]

Probably bc you charge to 100% every day which Tesla specifically says not to do unless you need it for long trips.

You can certainly do whatever you want as it is your car.

No, as you can see from my post, I’ve just decided to charge to 100%. I’ve been charging to 80%-90% for years now and it still loses. So, why not charge to 100%?

 

[Jeremy3292]

No, as you can see from my post, I’ve just decided to charge to 100%. I’ve been charging to 80%-90% for years now and it still loses. So, why not charge to 100%?

It will always lose some over time, but it will accelerate that timeline by charging to 100% regularly. Some of it could also just be the BMS needed to be calibrated. See this thread:

How I Recovered Half of my Battery's Lost Capacity

Like many others, I have been concerned with loss of 100% indicated battery range on one of my Model 3s. My P3D (build date 9/13/2018, delivery date 10/8/2018) had gotten down to 270.3 miles at 100% charge on January 20, 2020, at about 30,700 miles, which is a loss of 40.8 miles since the car...

teslamotorsclub.com

 

[AlanSubie4Life]

What am I worried about? Losing battery capacity? It already loses 1% a month.

Remember that calendar aging starts faster and then slows down. You’re actually doing fine at your current level of capacity loss. Charging to higher levels if you don’t need it will make calendar aging happen somewhat faster. If you need it, charge to 100% - it’s not a big deal especially in cold weather.

Capacity loss is part of owning an EV. If you have enough for your needs, it doesn’t matter. If you don’t have enough energy to make legs of a trip, it is a big deal. Increasing Supercharger density makes the second scenario less likely., and the capacity loss only mildly and incrementally increases trip time.

 

[AAKEE]

It will always lose some over time, but it will accelerate that timeline by charging to 100% regularly. Some of it could also just be the BMS needed to be calibrated.

If the charging level is kept below 90% (4.10V/cell) instead of 100%(4.20V/cell) the number of cycles is doubled / the wear cut in half.
By always charging to 100%, you double the wear from 90%.
Also, 90% causes more wear than 80 and so on.

 

[Jeremy3292]

Remember that calendar aging starts faster and then slows down. You’re actually doing fine at your current level of capacity loss. Charging to higher levels if you don’t need it will make calendar aging happen somewhat faster. If you need it, charge to 100% - it’s not a big deal especially in cold weather.

Capacity loss is part of owning an EV. If you have enough for your needs, it doesn’t matter. If you don’t have enough energy to make legs of a trip, it is a big deal. Increasing Supercharger density makes the second scenario less likely., and the capacity loss only mildly and incrementally increases trip time.

All other things being equal:

Scenario 1

• Always charge to max 100% for first 25k miles
• Always charge to max 50% next 25k miles (50k on odometer)

Scenario 2

• Always charge to max 50% for first 25k miles
• Always charge to max 100% next 25k miles (50k on odometer)

Which scenario will see more battery degradation? Or will it be the same?

 

[KenC]

All other things being equal:

Scenario 1

• Always charge to max 100% for first 25k miles
• Always charge to max 50% next 25k miles (50k on odometer)

Scenario 2

• Always charge to max 50% for first 25k miles
• Always charge to max 100% next 25k miles (50k on odometer)

Which scenario will see more battery degradation? Or will it be the same?

I was told there would be no math on the test.

 

[AlanSubie4Life]

All other things being equal:

Scenario 1

• Always charge to max 100% for first 25k miles
• Always charge to max 50% next 25k miles (50k on odometer)

Scenario 2

• Always charge to max 50% for first 25k miles
• Always charge to max 100% next 25k miles (50k on odometer)

Which scenario will see more battery degradation? Or will it be the same?

Yeah I have wondered about this too. Not sure if the vulnerability to calendar aging changes over time; even though we know that calendar aging slows over time at a given SOC, it might be “complicated.”

 

[AAKEE]

All other things being equal:

Scenario 1

• Always charge to max 100% for first 25k miles
• Always charge to max 50% next 25k miles (50k on odometer)

Scenario 2

• Always charge to max 50% for first 25k miles
• Always charge to max 100% next 25k miles (50k on odometer)

Which scenario will see more battery degradation? Or will it be the same?

I havent seen any research about that.
Probably not much difference.
First thing, its time that matters for calendar aging, not miles. 25K should be seen as a time period.
If any difference, scenario 2 might cause less degradation.
For the same SOC, we know that the calendar aging is worse initially( loss/time) and then lessens with the square root of time.
So using 100% early should probably wear more than using 100% later.

 

[AtticusRex]

It will always lose some over time, but it will accelerate that timeline by charging to 100% regularly. Some of it could also just be the BMS needed to be calibrated. See this thread:

How I Recovered Half of my Battery's Lost Capacity

Like many others, I have been concerned with loss of 100% indicated battery range on one of my Model 3s. My P3D (build date 9/13/2018, delivery date 10/8/2018) had gotten down to 270.3 miles at 100% charge on January 20, 2020, at about 30,700 miles, which is a loss of 40.8 miles since the car...

teslamotorsclub.com

Alright, you’ve convinced me. I’ll start charging to 80%. But, I’m still confused how to calibrate the BMS. Can you break that down for me?

 

[AAKEE]

I’ve been thinking about this and AAKEE’s posts a bit more. AAKEE pointed to a really good study here:

https://publications.lib.chalmers.se/records/fulltext/249356/249356.pdf

If you go down to section 5.2.1 (page 50), they have a similar chart/graph but out to 700 days or about 2 years. They tested calendar aging at 15%, 75%, and 90% SOC, and at 25C, 35C, and 45C.

Reading the graph, I’m concluding two things: 1. At 25C there was no appreciable difference in aging between 75% and 90% SoC. I.e. If you’re in a moderate climate, then the difference in battery calendar aging of 75% to 90% SoC is minimal to zero, maybe not worth dropping the charge down.

2. At 2 years in, the 15% SoC appeared to be at 98.5%-99% capacity, while 75% (and 90%) @ 25C appears to be 94.5%-95.0% capacity. Perhaps the inflection point is 55% as per the study above. If that’s linear over time - then at 10 years you’re looking at 5%-7.5% loss staying at 15% SoC, and maybe 25-27.5% loss at 75% or 90% SoC. (At 5 years the losses would be about 3% and 13% respectively).

It feels like if you’re going to keep the car for 5-6 years, I don’t see “3% degradation vs 13%” making much of a difference in resale value. For road trips at 5 years old, the 10% difference probably is tolerable too.

However, if you are keeping the car a lot longer - 10 years or more; at 10 years you probably also don’t havre too much difference in resale value - because 10 year newer EVs are going to be much better (and cheaper) in the battery department.. It’s going to be hard to trust that battery degradation really is low on a 10 year old vehicle even in the face of evidence, and “feeling” about a car tends to determine it’s price at that age. OTOH a road trip in a 10 year old EV with 25-30% range loss is huge vs. one with 5-7% range loss.

(I normally keep cars for 12-15 years myself (until they’re basically costing a lot of money and time to maintain), but with EVs improving so much every year especially over the next decade, I might revise that. Sell the 4-5 year old EV while it’s still worth something or similar. )

All this said I’m still thinking about limiting to 55% SoC based on the data above going forward to prevent further degradation of my 3…

That research report is good, with lot of good points. But the chemistry is NMC/LCO i think. Not the same.
To judge calendar aging when it comes to Tesla with Panasonic we should use research reports that did use NCA and preferably Panasonic NCA.
NCA calendar aging

This is representative, other research show about the same. Calendar aging lessens with time( square root of time). Also other similat tests with temp 10C and 15C show that lower temp is good for reducing calendar aging.

So, 25C for a year( about the 9.6 month line, as the car doesnt sleep all day) shows us that 70-80% SOC would cause about 5% for the first year.
To
Calculate after two years, square root of 2 times 5= About 7%.
Four years, square root of 4 times 5= 10%.

The “square root of time” seem to be accepted among most reserchers and I did see one research report trying to find a better formula, but actuallt it wasnt much better, as the batteries seems to more or less follow the square root principle.

Here is a graph with data points( the dots). The square root principle is clear.

For your calculations, Id say theres more calendar aging initially but it lessens with time:

 

[Jeremy3292]

Alright, you’ve convinced me. I’ll start charging to 80%. But, I’m still confused how to calibrate the BMS. Can you break that down for me?

You want to let your car “sleep” at different states of charge (SOC) - 32%, 85%, 61%, 90%, 44%, etc. This allows the car to take different readings at different SOC and to calibrate/rebalance itself. Sleep means no sentry mode, not charging, and no querying the car from the app (generally for at least three hours aka overnight).

Some people charge their car every single day to 80% or 90% regardless of how much they drove each day. So then the car never sleeps at a different state of charge so it only takes “readings” from the upper end of charge. This can cause the displayed rated miles to be inaccurate.

I personally do not charge every day as my commute round-trip uses only around 20% to 25%. So so my car will sit overnight at various SOC. Think back to the gasoline days - you wouldn’t go get gas when you had half a tank or 2/3 or 3/4 of a tank left so why would you charge your car with 50% or 66% or 75% remaining? It’s actually quite simple and makes logical sense when you think about it.

 

[AlanSubie4Life]

Alright, you’ve convinced me. I’ll start charging to 80%. But, I’m still confused how to calibrate the BMS. Can you break that down for me?

You can read the description just posted, but generally just don’t worry about it. For some people it makes a difference but those are typically exceptions. Your results are very typical so I wouldn’t bother. Just drive the car. It’ll figure it out, usually.

 

[AlanSubie4Life]

The “square root of time” seem to be accepted among most reserchers and I did see one research report trying to find a better formula,

Clearly begging for an experiment where they increase SOC after a couple years to answer the question above.

It would be pretty important to know that you could come out ahead if you really babied the battery for the first two years, and then stored it at a higher SOC later in life. I kind of doubt that it works that way (I think the higher SOC would re-accelerate the aging) but who knows.

 

[AlanSubie4Life]

No, for the LRs it's closer to 79kWh, with FPWN value of 82.1kWh. The 82.1kWh is just a nominal value; it's not clear why it is set that way or why vehicles don't get to that level. But it's just hard-coded anyway.

If you look in the 2021 thread there are a lot of SMT screen captures from both Performance vehicles (which get to 81kWh) and LR vehicles (which are usually closer to 79-80kWh).

One thing that complicates this is that for 2022 as has been pointed out elsewhere, they extracted 82.07kWh from the AWD, and 80.8kWh from the Performance. Yet, the degradation threshold is lower for the AWD at ~78.5-79kWh than it is for the Performance at ~80.7kWh (technically, TBD, but probably true). (https://dis.epa.gov/otaqpub/display_file.jsp?docid=54290&flag=1)

It's also not clear where the 2022 will end up on SMT. I'd be a bit surprised if it's STILL stuck at NFP 79-79.5kWh with 82.07kWh extracted in the EPA tests. However, as I have said elsewhere, there's always a bit of a mismatch between these SMT and EPA values, but a 3kWh difference is very much on the high side (within 1-1.5kWh is more typical).

Always mysterious though. It seems like the numbers are driven by engineering expectations rather than the numbers from any particular article (either the test article or owner vehicles). Might have been a particularly juicy AWD on the dyno.

There's four numbers to keep track of that always float around, which confuses people when quoting capacities:
1) FPWN (hard-coded)
2) NFP (vehicle-dependent)
3) EPA full pack extracted at nominal conditions (single article). (Cold conditions extraction is considerably lower)
4) Degradation threshold (applies to all vehicles of a given model year and trim level)

And then on top of that there's the fact that the usable capacity, to 0%, is 95.5% of the NFP.