Battery myths

[AAKEE]

There is a lot of myths* about lithium batteries floating around. Perhaps we can use this thread to kill or bust some of these, one by one?
There also might be statements that we ca not directly sort as facts or myth’s, lets find out.

For staters, which myth’s did you fall for? Try this quiz and see: Battery quiz



*) Myths: Statements that do not have any basis in science and which is not supported by research.

 

[AAKEE]

To start with low SOC, there is myth’s about low SOC being bad for the battery.

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

And this post: Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health


The posts above handle the cyclic aging of the battery.

Below calendar aging:

The usual Battery Degradation Topic

Hearing super conflicting advice about battery charging levels

Research data about calendar aging (from 2016 but doesnt differ from the latest reoprts: not 200 pages and somewhat easy to understand):
ShieldSquare Captcha

 

[dmwahl]

It would be interesting to compile information on failures of whole packs, which I suspect is mostly what people care about (the myth that cell failure causes most pack failures). For example contactor failures, failures due to water ingress, etc. I don't particularly care that my 85 pack is around 90% of original at 126k, even on long trips the only real impact is a little extra time at the supercharger. What I do care about is the sudden failures that are likely(?) non-cell failure.

The other one I happen to see floating around that would be worth some data is that frequent supercharger use degrades the pack faster. I haven't seen any evidence for that with my car, which according to SMT roughly 2/3 of the charging has been at superchargers.

 

[AAKEE]

The other one I happen to see floating around that would be worth some data is that frequent supercharger use degrades the pack faster. I haven't seen any evidence for that with my car, which according to SMT roughly 2/3 of the charging has been at superchargers.

If the batteries is sufficiently preheated the extra wear should not be that much.
Fast charging basically cause lithium plating and that can be reduced very much by having the battery preheated to at least 40 degrees C.

I think there was some model S that clearly took a hit from Supercharhing. This was begore my time so I do not know how Tesla did handle preheating back then?

The 2170’s in M3 seem to handle Supercharhing nice.

 

[3sr+buyer]

Regarding long term storage, it would be nice if the target charge could be set to 1%, so that the car can be stored long term plugged in without worrying about vampire drain leading to overdischarge.

 

[AAKEE]

Regarding long term storage, it would be nice if the target charge could be set to 1%, so that the car can be stored long term plugged in without worrying about vampire drain leading to overdischarge.

We need to account for the low voltage battery that ”drive the systems” when the car is asleep. I see about 1% SOC-loss per week (or less) during longer periods of parking. Sentry off, off course.
Tesla say to allow 1% per day when parking for longer times. That is high but I understand Tesla to set a margin to a broken 12V battery.
I would recommend following the Tesla advice to be surew to avoid broken LV batteries, but for myself I would am lower as I know my Tesla use less then 1% a week.

I left it at 33% before a US trip this summer. The 33% value did come from that a friend did catch the car at Tesla SC when I was packing form the trip. Not use to EVs I gave him safe number for the 300km drive home. If I had done the drive myself I had aimed lowe, like 15-20%.
It had 31% SOC after almost three weeks.

 

[Park2670]

Thanks for all the research you have done. Really appreciate the learnings and being able to apply them to my new car.

 

[DownshiftDre]

I will say it is thanks to AAKEE and reading and discussing with him in these forums that I was able to get a 12/13 on the quiz. The 100% SoC question was the one I got wrong and was a good trick question once I read the full answer commentary.

Thank you for creating this thread for the many others that can be educated in battery SoC management! Great idea!

 

[3sr+buyer]

I will say it is thanks to AAKEE and reading and discussing with him in these forums that I was able to get a 12/13 on the quiz. The 100% SoC question was the one I got wrong and was a good trick question once I read the full answer commentary.

Yes, that was a trick question, because it did not specify 100% versus what other state-of-charge. Someone who was thinking in comparison to <=55% for NCA or <=70% for LFP would answer opposite of the "correct" answer which assumes that the "incorrect" answer preferred 80-90%.

 

[shahryaran]

This question is misleading.
The question asking if 100% is damaging and not if 100% damages the same as 80%.
Reading the question implies "if 100% as a general rule damage the battery" but the answer is for a different question.
The answer is for a question asking "which SoC damages the most" in which answering 100% would be incorrect.

1. Is staying at 100% damage the battery: yes

2. Is staying at 80% damage the battery: yes

Both statements are correct. it's also true that both SoC have similar degradation and 80% may be even worse.
Even in the your answer feedback, you always compare 100% to 80%.

The question should be:
A lithium ion battery will gets damaged "more" at 100%.

In which the answer would be incorrect.

 

[AAKEE]

This question is misleading.
The question asking if 100% is damaging and not if 100% damages the same as 80%.
Reading the question implies "if 100% as a general rule damage the battery" but the answer is for a different question.
The answer is for a question asking "which SoC damages the most" in which answering 100% would be incorrect.

1. Is staying at 100% damage the battery: yes

2. Is staying at 80% damage the battery: yes

Both statements are correct. it's also true that both SoC have similar degradation and 80% may be even worse.
Even in the your answer feedback, you always compare 100% to 80%.

The question should be:
A lithium ion battery will gets damaged "more" at 100%.

In which the answer would be incorrect.

The truth is that the battery is not damaged at 100%.

If we went to a store and buy one of a few lihtium ion batteries, for example Panasonic NCR18650, that is closesly related to the Model S cells, if not exactly the same in some cases we will get a manual/specification stating this:
- Charge voltage 4.20V (compares to 100% in the tesla.)
- End of discharge 2.50V (way below 0% on the Tesla screen, and probably a bit below where the Tesla stops.
-There are numbers for which charging amps and discharging amps etc, but there are no precautions not to charge to 100% and leave them there. In fact, many of the areas of use of such lithium ion batteries involve charging them to 100% and leaving them there for long time.
My robot lawn mower always charge its 18650 full, and leaves it at that. It charge full after each lawning session, And sleeps for the rest of the 18h that day at 100%. For the winter, which is 6 months+ here, it should be charged full and stored.
I have some 18650 used for a battery pack for a FPV kit and they are charged full and then left in the pack for ”always”. After each use, I charge them full and leave them in the garage. Last time i measured the capacity it was on the track per the NMC research reports, I think these was 6 years. Time to check again, closing in to 8 years.
My Tesla model 3 cells(panasonic NCR2170), some of them, still are at 100% and closing in to 2 years. They are doing fine and actually has slightly less degradation than the 80% ones.

So, the per manufacturer spec, they do not get damaged by sitting at 100% but they degrade slowly as all batteries we know of so far.

The question is thought to give that understanding, maybe I need to look over the explanation?

 

[GSP]

There is a lot of myths* about lithium batteries floating around. Perhaps we can use this thread to kill or bust some of these, one by one?
There also might be statements that we ca not directly sort as facts or myth’s, lets find out.

For staters, which myth’s did you fall for? Try this quiz and see: Battery quiz



*) Myths: Statements that do not have any basis in science and which is not supported by research.

That battery quiz is quite good. I almost aced it, but I hit the wrong button by mistake on one.

I also selected 40% as the best SOC for long-term storage. The "correct" answer was 0%. Fair enough, as long as the battery is kept at 0%. That requires some sort of active storage where SOC is monitored and the battery is charged up when it falls below zero. I selected 40% because the battery should be good for 2 years or more unattended, as long as all loads have been removed.

GSP

 

[AAKEE]

That battery quiz is quite good. I almost aced it, but I hit the wrong button by mistake on one.

I also selected 40% as the best SOC for long-term storage. The "correct" answer was 0%. Fair enough, as long as the battery is kept at 0%. That requires some sort of active storage where SOC is monitored and the battery is charged up when it falls below zero. I selected 40% because the battery should be good for 2 years or more unattended, as long as all loads have been removed.

GSP

If you disconnect the battery as it will when reaching the mininum allowed voltage it will have a good margin to any overdischarge.

The research shows that the cells that was stopped discharging at the 0% stop discharge point increases in voltage during storage and the self discharge is very very low at 0%.

For the question ”What is the optimum storage SOC for a lithium ion battery to reduce the degradation from time ?” the answer would be 0% as it gives the lowest rate of calendar aging.

It can of course be discussed which is the smartest SOC number to use
The question above do not give a time window.

Tesla says 15-50% for storage of battery packs, and for those I guess the time window is set to possibly quite long.

 

[GSP]

@AAKEE - Thanks for your response. I've not seen data on self discharge rates vs. SOC. It is interesting to know the rate is lower at 0%. I guess that makes sense given the lower voltage.

However, most EV battery packs provide power to the cell balancing circuits directly from the cells. The current draw is very very low, but cannot be shut off, even with the service disconnect pulled. Leave a pack at 0% with cell voltage at spec sheet minimum, and I would expect the cells to be totally inert in a year or two.

GSP

 

[AAKEE]

@AAKEE - Thanks for your response. I've not seen data on self discharge rates vs. SOC. It is interesting to know the rate is lower at 0%. I guess that makes sense given the lower voltage.

However, most EV battery packs provide power to the cell balancing circuits directly from the cells. The current draw is very very low, but cannot be shut off, even with the service disconnect pulled. Leave a pack at 0% with cell voltage at spec sheet minimum, and I would expect the cells to be totally inert in a year or two.

GSP

Yes, we are not talking about or recommending to actually leave a pack at 0%. Just principles to know where the battery degrades and not.

BTW, I guess you know that 0% on the display is about 4.5% true so there is plenty of energy in that case.

I did not thinkt that the balancing was active all the time, and I do not think it is. I know I did read the patent application about that long time ago, but I do not remember the details.

Earlier I tested this with my M3P that had 4mV mostly at 55% SOC, charging 10% at the time slowly increased the imbalance, so it was 6mV at 85% a couple of hours after the charge was done.
Then I think I sat 93% or something, and about 2hrs after the charge was done, the imbalance was only 2 mV. It sure did look like the ”only balancing above X % or about 4.0V/cell.

For my MSP ’23, it seem to keep the slightly higher imbalance at 55% (like 10-12mV or so) but charging to 80-90% set in down to 2-4mV.
After three weeeks at 45-55% it still is at the same 12mV at rest and same SOC. So it doeasnt appear that any balancing is done.
I have seen really low imbalance values when driving, much lower than in my M3P.

 

[GSP]

@AAKEE - I did not mean to imply that cell balancing is always active, it certainly is not. However the logic chips that monitor cell voltage and run the software to activate the balance resistors, or not, are always drawing a small current. That current is much lower than used for active cell balancing.

GSP

 

[AAKEE]

@AAKEE - I did not mean to imply that cell balancing is always active, it certainly is not. However the logic chips that monitor cell voltage and run the software to activate the balance resistors, or not, are always drawing a small current. That current is much lower than used for active cell balancing.

GSP

I would guess that everything in the battery is off line (shut down) when the battery disconnects from the car after reaching the minimum voltage. BMS and all other things would in my mind be driven by the LV-system and that the HV battery itself is completely disconnected when the contactors open.

 

[GSP]

@AAKEE - Yes the main BMS computer is off, and it is usually powered by the 12 V system anyway, not the cells. HV contractors will open and HV load will be zero. However the chip that controls the balance resistors will still be powered by the cells. When balancing is complete it will go to a sleep state, but still draw some current.

Here is an example from Analog Devices, with LT8584 and LTC6804 chips powered fully or partially by direct connection to the cell groups they monitor and balance:

https://www.analog.com/media/en/technical-documentation/data-sheets/8584fb.pdf

"The part runs completely off of the cell which it is discharging, ... When the LT8584 is disabled, less than 20nA of total quiescent current is typically consumed from the battery."​

20 nA for each LT8584 is not much, but if I am reading the data sheet correctly the LTC6804 battery monitor chip takes 3.8 uA when in sleep mode.

See page 6:

https://www.analog.com/media/en/technical-documentation/data-sheets/680412fc.pdf

Tesla has a pretty advanced BMS on their newer cars, starting with the 2017 Model 3. It may operate differently than "typical" systems like I describe.

GSP

 

[Mayer]

A friend tagged me on Facebook with the following and I wondered whether it’s
factual at all, in anyone cares to reply:

This is a Tesla model Y battery. It takes up all of the space under the passenger compartment of the car.

To manufacture it you need:
--12 tons of rock for Lithium (can also be
extracted from sea water)
-- 5 tons of cobalt minerals (Most cobalt is made
as a byproduct of the processing of copper
and nickel ores. It is the most difficult material
to obtain for a battery and the most
expensive.)
-- 3 tons nickel ore
-- 12 tons of copper ore

You must move 250 tons of soil to obtain:
-- 26.5 pounds of Lithium
-- 30 pounds of nickel
-- 48.5 pounds of manganese
-- 15 pounds of cobalt

To manufacture the battery also requires:
-- 441 pounds of aluminum, steel and/or plastic
-- 112 pounds of graphite

The Caterpillar 994A is used for the earthmoving to obtain the essential minerals. It consumes 264 gallons of diesel in 12 hours.

Finally you get a “zero emissions” car.

Presently, the bulk of the necessary minerals for manufacturing the batteries come from China or Africa. Much of the labor for getting the minerals in Africa is done by children! If we buy electric cars, it's China who profits most!

BTW, this 2021 Tesla Model Y OEM battery (the cheapest Tesla battery) is currently for sale on the Internet for $4,999 not including shipping or installation. The battery weighs 1,000 pounds (you can imagine the shipping cost). The cost of Tesla batteries is:

Model 3 -- $14,000+ (Car MSRP $38,990)
Model Y -- $5,000–$5,500 (Car MSRP $47,740)
Model S -- $13,000–$20,000 (Car MSRP $74,990)
Model X -- $13,000+ (Car MSRP $79,990)

It takes SEVEN years for an electric car to reach net-zero CO2. The life expectancy of the batteries is 10 years (average). Only in the last three years do you begin to reduce your carbon footprint. Then the batteries have to be replaced and you lose all the gains you made in those three years.

The truth is far better than the fiction we are all being told.

 

[ItsNotAboutTheMoney]

A friend tagged me on Facebook with the following and I wondered whether it’s
factual at all, in anyone cares to reply:

This is a Tesla model Y battery. It takes up all of the space under the passenger compartment of the car.

To manufacture it you need:
--12 tons of rock for Lithium (can also be
extracted from sea water)
-- 5 tons of cobalt minerals (Most cobalt is made
as a byproduct of the processing of copper
and nickel ores. It is the most difficult material
to obtain for a battery and the most
expensive.)
-- 3 tons nickel ore
-- 12 tons of copper ore

You must move 250 tons of soil to obtain:
-- 26.5 pounds of Lithium
-- 30 pounds of nickel
-- 48.5 pounds of manganese
-- 15 pounds of cobalt

To manufacture the battery also requires:
-- 441 pounds of aluminum, steel and/or plastic
-- 112 pounds of graphite

The Caterpillar 994A is used for the earthmoving to obtain the essential minerals. It consumes 264 gallons of diesel in 12 hours.

Finally you get a “zero emissions” car.

Presently, the bulk of the necessary minerals for manufacturing the batteries come from China or Africa. Much of the labor for getting the minerals in Africa is done by children! If we buy electric cars, it's China who profits most!

BTW, this 2021 Tesla Model Y OEM battery (the cheapest Tesla battery) is currently for sale on the Internet for $4,999 not including shipping or installation. The battery weighs 1,000 pounds (you can imagine the shipping cost). The cost of Tesla batteries is:

Model 3 -- $14,000+ (Car MSRP $38,990)
Model Y -- $5,000–$5,500 (Car MSRP $47,740)
Model S -- $13,000–$20,000 (Car MSRP $74,990)
Model X -- $13,000+ (Car MSRP $79,990)

It takes SEVEN years for an electric car to reach net-zero CO2. The life expectancy of the batteries is 10 years (average). Only in the last three years do you begin to reduce your carbon footprint. Then the batteries have to be replaced and you lose all the gains you made in those three years.

The truth is far better than the fiction we are all being told.

Other studies usually put EVs as matching after 1 year due to the current grid mixes.

Besides pretending to be analytical while engaging in emotion, It also doesn't give any relative information. For example, it's only discussing batteries, without discussing comparable requirements for ICEV-specific components such as the engine, and, very significantly, without calculating for all the other common inputs to give an idea of how large the relative impact battery manufacturing has.

Then it just say "SEVEN years" without any calculation, even though the mass of other calculations on the topic show a much shorter period than that, maybe about a year. Lifecycle analysis summaries I've seen for vehicles have usually put fuel as 80% of overall lifecycle energy, so you can imagine how a more efficient vehicle can dramatically cut CO2 equivalence.

It reminds me of articles that circulated in the past year or two about disposal v recycling of PV, which deliberately ignored that recycling costs were a fraction of overall costs.