Long Term Battery Costs, Fears, and Serviceability

[MD70]

Great thread, I have been thinking about this subject as well.

For second hand buyers a great risk is what if your battery fails shortly after purchase.

I know that most vehicles if charged overnight or not supercharged every day can last for 2-300,000 miles or more. But some people will be the unlucky ones where the battery fails earlier.

I am thinking of buying a second hand model 3 which is in the warranty period but which I will keep as long as possible.

I think it would help if tesla publicly stated how much major repairs cost.

A UK member who had a faulty battery replaced had an invoice showing the battery cost to be £8,500.

So that at least gives me an idea of the cost if the worst happens.

Bringing down the battery cost and for owners to know the costs of major reports will help keep second hand values stable in my opinion.

 

[chinney]

There is a company called Re/Cell that is offering remanufactured battery packs for Model 3/Y for $8,000-10,000, including installation.

https://recell-ev.com/

The range of the remanufactured packs is somewhat limited though.

 

[AAKEE]

If I can get 300k miles and 13 years before I lose 25% capacity then I won’t need a battery change, I would get just a new vehicle at that point, unless you don’t think that’s possible

To be sure that the battery do not start deviating in excessive imbalance or such issue, you would like the degradation to stay below 20%.

20% degradation is the industry limit where lithium cells are considered consumed.
This is for a good reason, research test often show that they get non predictable after 20%.
As Tesla has large batteries eith normally low C-load during driving snd charging we might have well functioning packs with more degradation but as a rule of thumb I would strive to stay on the right side of 20%.

Like I’m not going to buy a 12-14k battery on a car that has 300k miles lol, the battery would cost more than the car is worth at that point

If you have new battery to a car you know is well taken care of, it might be a good choice. Even the motors aren’t that expensive if one needs a change.

 

[gottagofast]

If you have new battery to a car you know is well taken care of, it might be a good choice. Even the motors aren’t that expensive if one needs a change.

I still wouldn’t spend that much money on a car that old, I would be ready to buy a new car anyway, so is it impossible to stay below 20% degradation over 13-15 years?

 

[AAKEE]

I still wouldn’t spend that much money on a car that old, I would be ready to buy a new car anyway, so is it impossible to stay below 20% degradation over 13-15 years?

No, most probably not.

Hows your climate? Warm?

In a about average climate 80% daily (meaning average use, most often stay above 55%) would be about 20-21% calendar aging after 15 years an add maybe some for cyclic aging. 0.3 each year might be 24% after 15 years.

Using 55% or lower when thats enough and charge late-tactic’s will give about 11 calendar aging and maybe 3% cyclic aging, so 14% in total.

 

[gottagofast]

No, most probably not.

Hows your climate? Warm?

In a about average climate 80% daily (meaning average use, most often stay above 55%) would be about 20-21% calendar aging after 15 years an add maybe some for cyclic aging. 0.3 each year might be 24% after 15 years.

Using 55% or lower when thats enough and charge late-tactic’s will give about 11 calendar aging and maybe 3% cyclic aging, so 14% in total.

Atleast going by graphs and data It is possible to stay below 20% then over 15 years if you charge correctly from what you have looked at

even 24% degradation from charging to 80% daily might still be drivable but something that you would still want to try to avoid because it’s not guaranteed once your over 20%

 

[gottagofast]

Something disappointing to me is Tesla said these batteries were good for 1,500 cycles, I thought that meant 1,500 cycles period regardless of age.

So I thought a battery could be 20 or 30 years old and still be good if it has only been through 500 - 1,000 cycles

 

[Bouba]

I am hoping that replacement battery costs come down in the future so that when my current battery is reduced in range to a point where it is no longer all that practical for longer trips, it would be a reasonable option to replace it. Ideally the old battery could still do service in a solar installation or the like. The battery of my Model 3 LR is still doing pretty well after almost 4 years, but I can see a point maybe in another 4 years or so where I might want to replace it with a new one if the rest of the car is in good shape. My understanding is that the chassis and other basic components of Tesla vehicles are made to last for many, many years - and mine has been pretty much flawless so far - so it would be nice to hang on to it, but with, eventually, a fresh battery.

So far battery replacement costs are, I believe, too high for the above option to be all that realistic, but that could change over time. Perhaps there will be reputable 3rd-party providers that will fill the gap, if Tesla does not. Certainly a market for replacement batteries would be out there given the sheer number of Model 3/Y vehicles now on the road for which owners eventually will want to contemplate the possibility of eventual battery replacements. Also, there has been some talk of improved battery chemistry that soon may be coming that could someday make this a non-issue, as improved, long-life batteries will hopefully last the life of future vehicles without the degree of degradation that would make replacement an issue in most circumstances.

It’s also my hope that 3rd party battery repair companies will pop up and at reasonable prices and not just in the USA

 

[AAKEE]

Something disappointing to me is Tesla said these batteries were good for 1,500 cycles, I thought that meant 1,500 cycles period regardless of age.

So I thought a battery could be 20 or 30 years old and still be good if it has only been through 500 - 1,000 cycles

I havent heard/seen Tesla state that? Elon probably has, as he has made a lot of statements.

The focus is wrong being on miles or km driven, or cycles.

We know since long that panasonic NCA does about 800-1000 cycles if we do all of them in a row 100-0%.

Below 2018 model 3 Panasonic NCA 2170 cells. R50-100% means these was cycled between 100% and 50%. After 2500FCE which means 5000 cycles 100-50%, they did loose 20%.
2500FCE times 400 km = 1 million km. This is much better than the 1500 cycles you read/heard.
But these was performed on a row charge-discharge-charge-discharge (ususally a 15min brake between, I do not remember in this case).

If you add 15 years to this it would look much different. But 15 years a' 20k km is 300k km so 750FCE = About 10% in this case. But we do not go to the end points that often and mostly cycles are smaller than 50% DoD.

If you do smaller cycles and charge to 50-55%, you might end up like the green and blue line here, then these 750FCE during your 15 years would cause 4% cyclic aging in total.

I guess you have seen that calendar aging take a 7 times bigger bite on the battery for the first 10 years or so. This means tat if you focus on the cyclic aging only (getting information from someone stating 70 or 80% is the best for this) you might still have quite large calendar aging.

On the other hand, minimizing calendar aging will also make the cyclic aging go down (but in reality it is so small that it really doesnt matter for most.

 

[gottagofast]

I havent heard/seen Tesla state that? Elon probably has, as he has made a lot of statements.

The focus is wrong being on miles or km driven, or cycles.

We know since long that panasonic NCA does about 800-1000 cycles if we do all of them in a row 100-0%.

Below 2018 model 3 Panasonic NCA 2170 cells. R50-100% means these was cycled between 100% and 50%. After 2500FCE which means 5000 cycles 100-50%, they did loose 20%.
2500FCE times 400 km = 1 million km. This is much better than the 1500 cycles you read/heard.
But these was performed on a row charge-discharge-charge-discharge (ususally a 15min brake between, I do not remember in this case).

If you add 15 years to this it would look much different. But 15 years a' 20k km is 300k km so 750FCE = About 10% in this case. But we do not go to the end points that often and mostly cycles are smaller than 50% DoD.

View attachment 1024964

If you do smaller cycles and charge to 50-55%, you might end up like the green and blue line here, then these 750FCE during your 15 years would cause 4% cyclic aging in total. View attachment 1024965

I guess you have seen that calendar aging take a 7 times bigger bite on the battery for the first 10 years or so. This means tat if you focus on the cyclic aging only (getting information from someone stating 70 or 80% is the best for this) you might still have quite large calendar aging.

On the other hand, minimizing calendar aging will also make the cyclic aging go down (but in reality it is so small that it really doesnt matter for most.

It’s impossible for me to do smaller cycles when I drive 70 miles a day, I use 25% - 30% every day

I really shouldn’t charge over 55% for daily driving then going by that bottom graph

It looks like dropping below 30% is bad too

 

[AAKEE]

I really shouldn’t charge over 55% for daily driving then really going by that bottom graph

It looks like dropping below 30% is bad too

No, it is not.

It might look like that, but the picture fools you.
I did write about this several times the ladt weeks. There are pists with thourogh explanation.

In short, even 5-15%, which will be ~ 0-10% on our tesla displays (due to 4.5% buffer), will cause a very low annual cyclc degradation.

Calendar aging is less at very low SOC so the slight increase of cyclic aging (from very very low to very low) will be contered by reduced calendar aging.

 

[gottagofast]

No, it is not.

It might look like that, but the picture fools you.
I did write about this several times the ladt weeks. There are pists with thourogh explanation.

In short, even 5-15%, which will be ~ 0-10% on our tesla displays (due to 4.5% buffer), will cause a very low annual cyclc degradation.

Calendar aging is less at very low SOC so the slight increase of cyclic aging (from very very low to very low) will be contered by reduced calendar aging.

It looks like charging to 70% everyday isn’t too bad a compromise though it’s just slight more degradation

It’s like 4% more degradation than the perfect charging practice at 2,000 cycles

 

[AAKEE]

It looks like charging to 70% everyday isn’t too bad a compromise though it’s just slight more degradation

It’s like 4% more degradation than the perfect charging practice at 2,000 cycles

You’re thinking ”cyclic degradation” ?

Calendar aging above 60% is doubled.

The average EV might be driven 15-20K km a year? Thats 55km a day, 13% or so a day.

Charging to 70% means most people still have > 55% in the evd of the day. Because of this the calendar aging will be about the same as like having 80 or 90 or even 100%.


Another test and type of grsphic below:

So, do not focus to much on cyclic aging cause it might take ypu on the wrong turn.

 

[gottagofast]

You’re thinking ”cyclic degradation” ?

View attachment 1025004

Calendar aging above 60% is doubled.

The average EV might be driven 15-20K km a year? Thats 55km a day, 13% or so a day.

Charging to 70% means most people still have > 55% in the evd of the day. Because of this the calendar aging will be about the same as like having 80 or 90 or even 100%.


Another test and type of grsphic below:
View attachment 1025005

So, do not focus to much on cyclic aging cause it might take ypu on the wrong turn.

I use 25-30% a day though not 13%, so 70% might work a little different for me

Wait, your 2nd picture shows 100% charge is better than 70-90
70-90 is the worst for storage

 

[AAKEE]

I use 25-30% a day though not 13%, so 70% might work a little different for me

Wait, your 2nd picture shows 100% charge is better than 70-90
70-90 is the worst for storage

The research comes to slightly different results sometimes.
There are many tests that shows 80% is slightly worse than 100% and the ones that do not often show slight signs of it.

But as a general principle it better to think that above 55-60% the degradation is about the same for NCA. And that at high temperatures it is slightly worse at 100%.
The upper picture in the last post is good for remembering.

 

[SageBrush]

Occasionally, a battery pack can fail, requiring a battery replacement. This is where the real problem is.

I imagine that scenario is rare after warranty lapse, and view it as a good candidate for insurance. If the occurrence rate is 1% and replacement cost is $20k then the lifetime cost per car is $200 before company overhead and profit

My preferred option is for Tesla to build in the failure rate cost into the car price and warrant 300k miles, but it won't work because too many people will abuse the warranty. Perhaps a solution could be a hybrid scheme wherein Tesla insures everybody for 3/4 of the pack cost replacement

 

[chinney]

The battery situation with my Tesla Model 3 LR is a bit different that I had anticipated before I bought it 4 years ago. I had feared that to continue to accommodate more than just city driving (i.e, for long range use), I might need a new battery after about 5 years. That almost certainly will not be the case based on the limited degradation I have so far experienced. I was not all that familiar back then with just how good Tesla's batteries and battery management would be - although to be fair, my understanding is that batteries on its vehicles are lasting longer than even Tesla expected.

On the other hand, I had expected that battery prices would continue to come down and that significant advances would be made in battery technology so that any replacement battery I might buy would be at a reasonable price and might offer more range. The reality is that the slide in battery manufacturing costs that marked the previous decade did not continue over the last 4 years. Nor did the quick advances in battery technology. Costs have increased somewhat, if anything. And advances in technology, while not nothing, have been somewhat incremental. You can read stories almost every day in popular science and automotive newsfeeds suggesting that major advances in battery technology of various potential type are coming, but they are not here yet.

 

[gottagofast]

The battery situation with my Tesla Model 3 LR is a bit different that I had anticipated before I bought it 4 years ago. I had feared that to continue to accommodate more than just city driving (i.e, for long range use), I might need a new battery after about 5 years. That almost certainly will not be the case based on the limited degradation I have so far experienced. I was not all that familiar back then with just how good Tesla's batteries and battery management would be - although to be fair, my understanding is that batteries on its vehicles are lasting longer than even Tesla expected.

On the other hand, I had expected that battery prices would continue to come down and that significant advances would be made in battery technology so that any replacement battery I might buy would be at a reasonable price and might offer more range. The reality is that the slide in battery manufacturing costs that marked the previous decade did not continue over the last 4 years. Nor did the quick advances in battery technology. Costs have increased somewhat, if anything. And advances in technology, while not nothing, have been somewhat incremental. You can read stories almost every day in popular science and automotive newsfeeds suggesting that major advances in battery technology of various potential type are coming, but they are not here yet.

where have they lasted longer than Tesla expected? The oldest Tesla is only like 13 years old and I think that was expected, unexpected would be 20+ years

 

[fhteagle]

Extra data points for the discussion:

2013 Chevy Volt, in service since Sep 2012. ~99k miles on odometer, estimated 60/40 gasoline to electric miles.

16.5kWh gross, ~10.5 kWh usable at new. LiMNO2 chemistry in pouch format, not exactly what I'd call state of the art for longevity. Plugged in most nights but not all, and the pack has seen 110+F and -20F many many many times.

OBD reports ~12.7kWh gross now, and I see ~9.3kWh usable most of the time. Some of that useable reduction happened due to recall changing the bottom buffer values a bit.

Junkyard packs are running ~$1900 plus logistics to get it to you, if you know what you're doing you could swap packs in an afternoon, or exchange modules in a couple of days of careful labor. 3rd party rebuilt packs using matched modules can be installed for $5-8k from a couple of different places. Stealerships quoting $15k-20k are trying to scare people into buying new.

If we want to mitigate replacement costs, standardizing pack and/or module size, shape, capacity, voltage, etc is the most important thing in my opinion. Major OEMs don't engineer a whole new engine for every ICEV nameplate. The sooner that same thinking of using the same pack bits across a whole brand or auto group actually starts happening, the better off we all will be.

 

[gottagofast]

Extra data points for the discussion:

2013 Chevy Volt, in service since Sep 2012. ~99k miles on odometer, estimated 60/40 gasoline to electric miles.

16.5kWh gross, ~10.5 kWh usable at new. LiMNO2 chemistry in pouch format, not exactly what I'd call state of the art for longevity. Plugged in most nights but not all, and the pack has seen 110+F and -20F many many many times.

OBD reports ~12.7kWh gross now, and I see ~9.3kWh usable most of the time. Some of that useable reduction happened due to recall changing the bottom buffer values a bit.

Junkyard packs are running ~$1900 plus logistics to get it to you, if you know what you're doing you could swap packs in an afternoon, or exchange modules in a couple of days of careful labor. 3rd party rebuilt packs using matched modules can be installed for $5-8k from a couple of different places. Stealerships quoting $15k-20k are trying to scare people into buying new.

If we want to mitigate replacement costs, standardizing pack and/or module size, shape, capacity, voltage, etc is the most important thing in my opinion. Major OEMs don't engineer a whole new engine for every ICEV nameplate. The sooner that same thinking of using the same pack bits across a whole brand or auto group actually starts happening, the better off we all will be.

That’s good results for that pack, and was it even charged carefully or was it charged to 100% on a regular basis and at 0% at regular basis?