LFP to 100% Each Week—Assumptions?

[stopcrazypp]

China have been using the LFP battery pack for awhile. The data so far, does suggest the LFP degrades slower than the NCA.

I had a 2019 M3 and I babied that battery. My SOC went from 240 to 218 after just 20,000 miles. Not good. I am expecting my new M3 LFP to have much less degradation.

This is expected behavior of LFP vs NCA. This is well established in battery data sheets and well known by those of us that have been following batteries a long while. LFP generally has much higher cycle life and tolerates cycling much better. That doesn't mean however 100% is actually good for it and 50% has no advantages. It just takes longer to show up due to higher cycle life.

BTW do you have links to the data gathered for China vs how it tracks with NCA?

 

[stopcrazypp]

Again, just speculation. LFP RWDs have been on sale and in production for ~ 2 years now. Where's the data showing less range loss due to a specific charging strategy?

It's informed speculation. Previous general degradation of NCA (as in what SOCs result in better outcomes) tracked with what battery datasheets and research showed. Less than 2 years in ownership is not enough data to say there is no contribution, as mentioned (especially given higher cycle life in LFPs in the first place, which would extend the amount of time it takes to observe the differences). Point of the matter is, given raw CATL LFP cells have been demonstrated to like being stored at lower SOCs more than they do at 100%, it is highly likely that it's the same in the LFP packs Tesla is using (there hasn't been any indication Tesla is using any new advanced magical chemistry from them).

The only reason why this is even being discussed in regards to LFPs is the contradictory recommendation by Tesla to keep the car 100% charged, and the only reasoning for that recommendation is because of the low SOC estimate problem talked about previously. It's not because it's "good" for the battery cells to be charged at 100%.

And as others repeated, people should do whatever they feel is convenient for them, just be aware of the general cell degradation chemistry. For myself personally (with an NCA car), charging to 100% for regular use is actually an inconvenience, because it reduces my regen capabilities. I'm much happier at 70%.

 

[Niroc]

It's informed speculation. Previous general degradation of NCA (as in what SOCs result in better outcomes) tracked with what battery datasheets and research showed. Less than 2 years in ownership is not enough data to say there is no contribution, as mentioned (especially given higher cycle life in LFPs in the first place, which would extend the amount of time it takes to observe the differences). Point of the matter is, given raw CATL LFP cells have been demonstrated to like being stored at lower SOCs more than they do at 100%, it is highly likely that it's the same in the LFP packs Tesla is using (there hasn't been any indication Tesla is using any new advanced magical chemistry from them).

The only reason why this is even being discussed in regards to LFPs is the contradictory recommendation by Tesla to keep the car 100% charged, and the only reasoning for that recommendation is because of the low SOC estimate problem talked about previously. It's not because it's "good" for the battery cells to be charged at 100%.

And as others repeated, people should do whatever they feel is convenient for them, just be aware of the general cell degradation chemistry. For myself personally (with an NCA car), charging to 100% for regular use is actually an inconvenience, because it reduces my regen capabilities. I'm much happier at 70%.

The LFP battery from CATL is the third generation, I believe. I can only assume they improved not only density but the stability as well. I think we all know the LFP battery pack does degrade. But it is comforting to know that the LFP has the ability to last for quite some time

 

[stopcrazypp]

China have been using the LFP battery pack for awhile. The data so far, does suggest the LFP degrades slower than the NCA.

I had a 2019 M3 and I babied that battery. My SOC went from 240 to 218 after just 20,000 miles. Not good. I am expecting my new M3 LFP to have much less degradation.

To go back in memory lane, as a sanity check, here is a post I made in 2011 (more than a decade ago!) based on the NCR18650A cell datasheets which were the closest to the expected NCA cells in Model S (which wasn't even released yet).
Panasonic cells for Model S
NCR18650A were mass produced starting at end of 2009:
Panasonic Starts Mass-Production of High-Capacity 3.1 Ah 18650 Li-ion Cells

Here's the numbers scaled to 240 miles max (instead of the 300 miles I used back then) and using linear approximation to get miles (I added the ~300 cycle knee point, subtracted the 5 cycle rapid degradation stage at start, added 20k mile point).

full cycle	total miles	mAh	range (mi)
0​	0​	3000​	240​
69​	15870​	2750​	220​
89​	20000​	2708​	217​
185​	40230​	2500​	200​
295​	62010​	2450​	196​
458​	92980​	2300​	184​

You see even with almost a decade of cell improvements, your NCA degradation result tracked pretty well with the actual datasheets of similar cells. My own degradation on my 2021 SR+ also tracks well (from 263 miles brand new to ~250 miles at around 10k miles). Basically the general trends tend to hold pretty decently with the data sheets.

It's stuff like this why I don't expect the general trends to be too far off from datasheets in terms of how cells degrade.

 

[Niroc]

To go back in memory lane, as a sanity check, here is a post I made in 2011 (more than a decade ago!) based on the NCR18650A cell datasheets which were the closest to the expected NCA cells in Model S (which wasn't even released yet).
Panasonic cells for Model S
NCR18650A were mass produced starting at end of 2009:
Panasonic Starts Mass-Production of High-Capacity 3.1 Ah 18650 Li-ion Cells

Here's the numbers scaled to 240 miles max (instead of the 300 miles I used back then) and using linear approximation to get miles (I added the ~300 cycle knee point, subtracted the 5 cycle rapid degradation stage at start, added 20k mile point).

full cycle	total miles	mAh	range (mi)
0​	0​	3000​	240​
69​	15870​	2750​	220​
89​	20000​	2708​	217​
185​	40230​	2500​	200​
295​	62010​	2450​	196​
458​	92980​	2300​	184​

You see even with almost a decade of cell improvements, your NCA degradation result tracked pretty well with the actual datasheets of similar cells. My own degradation on my 2021 SR+ also tracks well (from 263 miles brand new to ~250 miles at around 10k miles). Basically the general trends tend to hold pretty decently with the data sheets.

It's stuff like this why I don't expect the general trends to be too far off from datasheets in terms of how cells degrade.

Nice! But didnt the earlier model S use a different cell format than the current 2170?

 

[PianoAl]

I had a 2019 M3 and I babied that battery. My SOC went from 240 to 218 after just 20,000 miles

Only 20,000 miles, but four years of calendar degradation, right? Degradation of 10% isn't bad given that and is what we're told to expect, so you shouldn't be bummed.

 

[stopcrazypp]

Nice! But didnt the earlier model S use a different cell format than the current 2170?

Yeah they used 18650 cylindrical cells, just like the datasheet cells, but the difference between that and 2170 is only a cell format difference, it generally should have little to no contribution to degradation rate, given degradation is largely based on the cell chemistry (which is the same or similar) for similar cell formats (both are metal cylinders).

The LFP cells are prismatic (like bricks instead of cylinders).

 

[Niroc]

Only 20,000 miles, but four years of calendar degradation, right? Degradation of 10% isn't bad given that and is what we're told to expect, so you shouldn't be bummed.

I am not bummed as I sold that car last year for money than what I paid for it.

 

[Niroc]

Yeah they used 18650 cylindrical cells, just like the datasheet cells, but the difference between that and 2170 is only a cell format difference, it generally should have little to no contribution to degradation rate, given degradation is largely based on the cell chemistry (which is the same or similar) for similar cell formats (both are metal cylinders).

The LFP cells are prismatic (like bricks instead of cylinders).

Yep. Munro had a review of the LFP battery pack. He was super impressed with CATL's work.

 

[DrChaos]

I haven't seen any evidence that charging TM3 LFP batteries according to the recommendations causes the battery to degrade faster.

Tesla and CATL have not changed the facts of chemistry. Calendar degradation is lower at lower state of charge, and calendar degradation is the primary mechanism unless driven very heavily like daily commercial use.

Tesla's recommendations are to reduce service calls and avoid PR problems. LFP degradation generally is better than other chemistries to begin with, but a lower daily SOC with once a week 100% will result in better lifetime.

 

[DrChaos]

China have been using the LFP battery pack for awhile. The data so far, does suggest the LFP degrades slower than the NCA.

I had a 2019 M3 and I babied that battery. My SOC went from 240 to 218 after just 20,000 miles. Not good. I am expecting my new M3 LFP to have much less degradation.

What specifically do you mean by "babied that battery"? I'm at 2.7% degradation after 9 months, usually keeping SOC limit at 50%.

 

[DuncanM]

Tesla and CATL have not changed the facts of chemistry. Calendar degradation is lower at lower state of charge, and calendar degradation is the primary mechanism unless driven very heavily like daily commercial use.

Tesla's recommendations are to reduce service calls and avoid PR problems. LFP degradation generally is better than other chemistries to begin with, but a lower daily SOC with once a week 100% will result in better lifetime.

Again, it should be clear from owner experiences and data that a particular charging strategy result in less degradation...so where's the data?

 

[DrChaos]

Again, it should be clear from owner experiences and data that a particular charging strategy result in less degradation...so where's the data?

We won't know for years because the main mechanism is calendar aging and LFP Teslas are only recently developed. LFP degrades less to begin with, and finally almost nobody (in the general population) sets the daily charge limit low enough (60% or less regularly) for long enough a time for a significant effect to be visible. Even the statistical apps for Tesla nerds segregate battery degradation by distance travelled and not age or average state of charge or number of/depth of cycles, and there's a big noise effect as a result. A more comprehensive statistical analysis would try to model all those multivariate effects.

So we have to go with controlled scientific research results on basic cells. I'm following AAKEE's fantastic analysis and setting my NCA charge limit to 50%, and have very little degradation (which all came in hot summer months anyway, as predicted by science), as does he.

 

[Whoze]

What they forgot to include in the manual:

We recommend this even though it will cause your battery to degrade faster because we need to keep the BMS well calibrated. Otherwise people might be stranded, resulting in a front page story about how Teslas run out of charge unexpectedly.

Thanks for the reply. I read up on a related thread that posted a battery discharge curve, and a discussion about the BMS. The book does say to charge it to 100% at least once a month. This would be consistent with the BMS calibration. In the same way, we were advised to fully discharge the older cell phone Li battery once a month to ensure the battery monitor could accurately charge it to the maximum capacity. As I do 2 long 250-mile commutes a month, and the rest are mostly 5 to 10-mile trips around town, it would seem to make sense for me to charge it to 100% before and after the long trips, and not to worry too much about the around-town drives. TBH, It rarely drops below 70% around town even when I don't charge it for the week.
It does blow my mind that nobody mades headline news when a gasoline car runs out of juice, considering a standard Ford Model T had a range of 155 miles!

 

[Whoze]

Same comment made a few months ago, but hundreds of thousands of vehicles and millions of miles is quite irrelevant if each individual pack have not accumulated enough cycles and calendar wear. To put it more clearly, here's two scenarios, say you have 100 million miles of data:
1) 1000 cars traveling 100k miles
2) 100k cars traveling 1000 miles

#1 will tell you a lot about long term wear. #2 will tell you almost nothing. Currently the LFP fleet is a lot closer to #2 than #1. There's probably a while before there are many 100k mile LFPs on the road.

@AAKEE already posted the data on the basic chemistry of LFP that shows 50% will do better than 100% for storage. Tesla's statements don't contradict that and the only reasoning that Tesla have presented to recommend the 100% is because the LFP BMS has inaccurate capacity estimates at low SOCs below ~10% (increasing the risk of stranding). Besides from that article, I previously linked a tweet thread by Elon that makes it much more clear:

https://twitter.com/x/status/1498859400525750272

I Picked up my Model 3 today. How can I tell if has LFP battery pack?

Basically it's a double whammy. The LFP packs + the 16V lithium LV battery means that if the HV shuts off unexpectedly (instead of a gradual shutoff), the LV battery can die very quickly (given it has much lower capacity of 6.9 Ah vs 45 Ah for the lead acids). This is one reason why I actually may prefer the lead acid (other than also easier to find a generic replacement if it dies).

I would also like to add that when the red light flashes on a gasoline car, the remaining range can mean anything. If the remaining miles displayed were to go by, I have driven cars with antimatter energy as the fuel zeroed out 10 miles ago, sometimes more, and other times I need to pull out the can. So, the easy fix is to have the batteries charge to "110%" but display it as 100%. So you will have a reserve of 10%. That seems to be what ICE vehicles do anyway.
But I guess people will start complaining about "Phantom Battery" (I coined that phrase - it's mine! haha).

 

[Niroc]

We won't know for years because the main mechanism is calendar aging and LFP Teslas are only recently developed. LFP degrades less to begin with, and finally almost nobody (in the general population) sets the daily charge limit low enough (60% or less regularly) for long enough a time for a significant effect to be visible. Even the statistical apps for Tesla nerds segregate battery degradation by distance travelled and not age or average state of charge or number of/depth of cycles, and there's a big noise effect as a result. A more comprehensive statistical analysis would try to model all those multivariate effects.

So we have to go with controlled scientific research results on basic cells. I'm following AAKEE's fantastic analysis and setting my NCA charge limit to 50%, and have very little degradation (which all came in hot summer months anyway, as predicted by science), as does he.

Tesla didn't develop anything concerning LFP. CATL did. And, thank goodness for that. CATL creates the battery pack and ships it to Tesla. CATL are the premiere experts in battery technology. I think their LFP technology will eclipse Tesla's 4680 battery production as they will keep improving on the prismatic LFP design. Of course, I am a bit skeptical on 4680 so I am a bit of Missouri on that. (Show me)

I had to chuckle at your "predicted by science". There is actual science and today's scientist. Big distinction. Modern scientists are one the biggest scam artists on this planet. Give them a big grant and a title and they will say anything you want to hear.

 

[PianoAl]

Apparently the BMS currently bases its prediction only on voltage (yes?).

However, the car knows all of the charging and driving history, temperature history, and can reestimate past SOCs based on how long it took to recharge. Perhaps in the future, when we have more data, all of this information can be used to get a better estimate.

 

[AAKEE]

I had a 2019 M3 and I babied that battery.

How did you baby the battery?

 

[AAKEE]

Apparently the BMS currently bases its prediction only on voltage (yes?).

The ”real” SOC is measured in voltage.
Each SOC number, like 50 or 70 or 100% relates to a cell voltage when the car is at rest, HV Battery contactors open.

During a drive, the last know SOC is the initial SOC and the shown SOC during the drive comes from initial SOC minus the used energy (kWh). The calculated SOC is approximate, as there is faults like wrong capacity estimate etc.
After the drive when the HV Batt is at rest, the true SOC can be measured, and the on screen SOC adjusted.

For the LFP, the voltage curve is so flat that the OVC reading might lead to a wrong SOC number.

 

[Niroc]

How did you baby the battery?

I rarely charged to 100% I never dropped below 30% and hardly charged over 80%.