Fun with chromebook battery / how Google/Acer hate my battery

[magic]

Many years ago I bought an Acer chromebook, wanting a simple and lightweight laptop with long battery life and also an ARM machine to play with.

At some point I sorta stopped using it and tossed it into storage. That's where the fun began.

Firstly, I noticed that it discharges the battery fairly fast. It was easily 50% after a few months.
Secondly, when the battery goes very low (I haven't determined how low exactly) the battery/power LED starts to blink even when the machine is "turned off"
It appears that in this state, it discharges itself down to the hard protection in the battery itself, at 2.5V per cell. Job well done
When I reconnected it yesterday, the battery didn't even want to charge initially, only started on the second or third try.

I have previously stored unpowered x86 laptops with batteries in them for months too and never seen similar problems.

[Junaid_raza]

The issues you are experiencing with the Acer Chromebook are likely due to the type of battery used in the device. Chromebooks typically use Lithium-ion batteries, which can discharge over time, especially if stored in a partially charged state. Additionally, Lithium-ion batteries can be damaged if they discharge below a certain voltage, which can cause the battery to lose capacity or even fail completely.

Compared to older x86 laptops, which may have used different battery chemistries or had larger batteries, the Lithium-ion battery in your Chromebook is more susceptible to these types of issues. To prevent battery discharge and damage, it's generally recommended to store Lithium-ion batteries in a cool, dry place with a charge level between 40-60%.

[Someone]

Compared to older x86 laptops, which may have used different battery chemistries

Lol, ChatGPT answer right there. Lithium-Ion has been the dominant battery chemistry in portable electronics well before x86 was a term (or widespread internet use).

[SiliconWizard]

Compared to older x86 laptops, which may have used different battery chemistries

Lol, ChatGPT answer right there. Lithium-Ion has been the dominant battery chemistry in portable electronics well before x86 was a term (or widespread internet use).

Are you sure?
I don't know exactly when the term "x86" alone became popular - I'd say probably around when the 486 got released (1989), when there were enough generations of the x86 instruction set that it warranted a nickname. But it may have appeared even earlier. With the 486, that's also the time when CPU clones appeared with 'x86' in their names, such as the 5x86 series from AMD. Just a guess though.

I obviously don't know the full history of laptops, but one related reference I have is the first IBM Thinkpad (1992), the 700. So, 1992, a 486 CPU and... a NiMH battery.

Li-ion batteries in laptops started appearing much later. "x86" had probably been a nickname for the x86 architecture for a while at this point. Just a detail though, but you may have overestimated how early Li-ion batteries became pervasive in portable devices.

[alm]

Early NiMH cells (and before that NiCd) had a much higher self discharge than Li-ion, so the answer still doesn't make sense.

To me it sounds like either aging of the battery, where a cell might have been disabled by the protection circuit, or maybe just bad design and high current draw while turned off.

[RoGeorge]

Firstly, I noticed that it discharges the battery fairly fast. It was easily 50% after a few months.
Secondly, when the battery goes very low (I haven't determined how low exactly) the battery/power LED starts to blink even when the machine is "turned off"
It appears that in this state, it discharges itself down to the hard protection in the battery itself, at 2.5V per cell. Job well done
When I reconnected it yesterday, the battery didn't even want to charge initially, only started on the second or third try.

I have previously stored unpowered x86 laptops with batteries in them for months too and never seen similar problems.

- Li-Ion self discharge rate is very small.  At the same time, Li-ion doesn't cope well with full charge.  Some Li-Ion manufacturers specify in their datasheets to not store fully charged battery for long term.  I've read some smart battery packs for drones will self discharge to about 50% if left 100% charged and unused for too long.

- there's nothing wrong if the LED blinks, the BMS (Battery Management System) will disconnect the battery completely when the voltage goes too low.

- when completely discharged, after the BMS cut out the power, you need to plug the charger and wait.  Can take a few minutes until visible charging happens.  At first the charging current is very gentle, so to not damage the battery.  Plugging and unplugging doesn't help, while you plugged it 3 times, you just complete the time needed to wait.  Plug once and let it be.  It will start charging by itself, just not immediately.

- older laptops were not so careful with their batteries, but when you need to maximize battery life, like in a Chromebook, then you'll need to apply all the good practices, even when that will make users think you hate their batteries. 

[Someone]

Compared to older x86 laptops, which may have used different battery chemistries

Lol, ChatGPT answer right there. Lithium-Ion has been the dominant battery chemistry in portable electronics well before x86 was a term (or widespread internet use).

Are you sure?
I don't know exactly when the term "x86" alone became popular - I'd say probably around when the 486 got released (1989), when there were enough generations of the x86 instruction set that it warranted a nickname. But it may have appeared even earlier. With the 486, that's also the time when CPU clones appeared with 'x86' in their names, such as the 5x86 series from AMD. Just a guess though.

I obviously don't know the full history of laptops, but one related reference I have is the first IBM Thinkpad (1992), the 700. So, 1992, a 486 CPU and... a NiMH battery.

Li-ion batteries in laptops started appearing much later. "x86" had probably been a nickname for the x86 architecture for a while at this point. Just a detail though, but you may have overestimated how early Li-ion batteries became pervasive in portable devices.

The usage I am familliar with for x86 was once the architectures became similar enough/interchangeable that the x was irrelevant. 186-286-386-486 were all significantly different, "pentium" 586 a smaller step again and beyond that software compatibility was pretty stable. Intel put out P6 in 1997 by which point there were very very few laptops using battery chemistry other than lithium ion.

Talking in a general sense about batteries from 30 year old computers is hardly relevant to contemporary computer and battery use. How many portable computers of that era are still in use? People using such antiquated devices are probably well aware of the details. We dont give people talks about carburettors when the question is why their modern car is having problems!

[magic]

Lol, ChatGPT answer right there.

This guy is such a moron it's not even funny. Using ChatGPT to farm posts on EEVblog forum and then shamelessly posting links like below to the very same forum

https:// sturdy business . com/does-ai-content-rank-on-google/

Check out his post history, besides self-promotion of his website it's all AI generated gibberish or talking about AI. Very stealthy

edit
Too late, Halcyon took care of it

[magic]

- Li-Ion self discharge rate is very small.  At the same time, Li-ion doesn't cope well with full charge.  Some Li-Ion manufacturers specify in their datasheets to not store fully charged battery for long term.  I've read some smart battery packs for drones will self discharge to about 50% if left 100% charged and unused for too long.

Show me those who recommend long term storage at 2.5V per cell (not to the same as 3.5V ).

- there's nothing wrong if the LED blinks, the BMS (Battery Management System) will disconnect the battery completely when the voltage goes too low.

The BMS cuts out at something like the absolute minimum rating of Li-ion cells, and from there it could only keep getting worse due to self-discharge and BMS drain.

Yes, it's an idiotic design. It is simply impossible to store this machine long-term in battery-friendly conditions (near 50% IIRC) without disassembling it and removing the battery.

- when completely discharged, after the BMS cut out the power, you need to plug the charger and wait.  Can take a few minutes until visible charging happens.  At first the charging current is very gentle, so to not damage the battery.  Plugging and unplugging doesn't help, while you plugged it 3 times, you just complete the time needed to wait.  Plug once and let it be.  It will start charging by itself, just not immediately.

Yes, the pack has recovered. But the BMS was initially reporting it as "failed" to the software and the laptop wouldn't last even a second without external power, so I was pissed and wrote this rant

- older laptops were not so careful with their batteries, but when you need to maximize battery life, like in a Chromebook, then you'll need to apply all the good practices, even when that will make users think you hate their batteries. 

There is virtually no energy to be gained overdischarging below 3.0V (cell voltage drops like a stone below that point) and that's the usual turn-off threshold as far as I understand, 2.5V being more like a definite point of no return that you never want to cross even under failure conditions.

[RoGeorge]

Indeed, and I was ranting at your rant. 

About the blinking LED, it would need very little energy to blink.  My guess is 0.1mA on average would be enough (10mA when on and 1:100 blinking ratio).  What the laptop will consume in 20 seconds, will keep the LED blinking for days.  Now put the designer shoes to decide this: 
   - prolonging the run of the laptop with 20 more seconds, or
   - blinking that LED for 2 more days, to remind the user that the battery needs charging ASAP
Which one would you choose?

[magic]

It was blinking when it was turned off until the BMS cut it off.

A decently designed laptop, instead of blinking for 2 days, could stay out of overdischarge and even keep the RTC running for 20 if not 200 days - that's the tradeoff being made.