Effect of voltage on battery wear (Li-Ion / phones)

I have been using an app that tracks the amount of cycles my phone battery makes and can help to lower the amount of cycles (gives a warning if the battery level is 'too high'). This can extend the lifetime.

Normally I dislike those so called battery-apps, but this one seems to actually have scientific backup and no ridiculous promises.

Anyway, on the app's website they have this article that explains it: https://accubattery.zendesk.com/hc/en-us/articles/210224725-Charging-research-and-methodology

I wanted to share this because perhaps it can help your battery/phone life.

If you don't want to click the link: this is the main point:

Choi 2002 tested how charging to different voltages affect cycle performance, and the data is dramatic: for every 0.10V increase, the life time of the battery is cut to about half, which Asakura 2003 also concluded with "about 0.1 V increase in charging voltage also cut the cell life in half."

Click to expand...

^^^ That 2002 Choi study (and many others) have been discussed here at length in the past. You can find more in-depth (and more recent) information in the discussions here (e.g. the effects of balancing partial discharges around 50% SOC).

Gauss163 said:

...the effects of balancing partial discharges around 50% SOC.

Click to expand...

Could you please explain 'balancing partial discharges'? I am getting a new phone and would like to maximize battery life (within reason).
So far I try to stop charging at 80% and start charging at 20%, roughly.
Do you think an app like that could make a significant difference?

^^^ Your 20-80% cycles are already perfectly balanced around 50% SOC, namely from 30% below to 30% above. Studies have shown that generally this minimizes the wear since it minimizes the time that the cell spends as extreme voltages.

I think when it comes to phones, you're probably going to replace the phone before the battery dies from staying at high-voltage too much. So, I wouldn't obsess about it too much for electronics with designed obsolescence.

And, frankly, I wouldn't want to rely on a phone with only 20% charge. Charge it up well before that.

As an example:

On 27 January 2017 I installed the app on a brand new phone and have been using it all this time.

The charge "full warning" is at 80% which I only missed around ~25 times. I don't care about low charge so it was often somewhere between 3-30% before hooking it up on the charger which has Turbo.

Turbo charging means on average around 66% charge is added per hour. I think >90% charging was done on Turbo.

The phone has done in total 519 charging sessions with a total of 831.449 mAh (with a 3.000mAh battery).

The estimated total charge capacity by the app is still 108%. Which I think is too high already from the start, but it hasn't dropped, so the battery is likely still close to the start capacity.

I can provide images as proof if needed.

Thank you Gauss.

And yes, depending on anticipated use, it would seem better to charge before reaching 20% but that does give some guideline to lessen the phone's use if not around charging capability.
Also just seems 'best use' to not charge to 100% if it is convenient and practical, because of documented research indicating extreme states of charge causes increased cell degradation; namely not leaving the battery at 0% for an extended period, and not keeping the phone plugged in charging mode all the time. I understand that avoiding use of the phone while charging is better because the battery will get hot? And also not leaving the phone in a very hot environment for very long.

Hoping the phone I get will last a long time; not planning on replacing it any time soon. We still have an early iPad 2!

NiMHi said:

[...] The phone has done in total 519 charging sessions with a total of 831.449 mAh (with a 3.000mAh battery). The estimated total charge capacity by the app is still 108%. Which I think is too high already from the start, but it hasn't dropped, so the battery is likely still close to the start capacity.

Click to expand...

So your cycles have average depth of about 53% = 831/(519*3), so you're using the region 27% - 80% SOC, and you've done 277 = 831/3 equivalent full cycles. That amount of cycling will surely cause some nontrivial capacity loss - though less than if you did it with deeper cycles. It is unlikely that the battery is still close to the start capacity, i.e. the app is probably not very accurate in this regard. Why don't you do a full cycle and see what it reports for the current capacity?

You should do a full cycle (0% to 100%) on your phone every once in awhile anyway, to reset its internal battery indicator. Otherwise, the battery indicator will gradually get more and more inaccurate.

I did a charge test from 2% to 99% (2h 20m charging with "turbo mode", did not use the phone). Removed charger with still 104mA charge current and (charging) voltage of 4,341 mV.

Waiting a minute or so after disconnecting cable showed 4,275mV for the batery (with -136mA). The estimated capacity for my 3,000 mAh battery from this charge is 2,849.2 mAh.

I'm starting to think that I never got a 3000mAh battery in my Moto G4 Plus. All the other large charging sessions* I did also show around 2800 mAh. Check the graph below from the app to show what I mean (I added the red line).



* The graph below are all the charge sessions including a few big ones to determine the best estimated capacity:

I virtually never charge my tablet or phone over 80% (battery built in) or discharge below 30%, I only charge to 100% for battery calibration reasons, and i have only done this 5-8 time in nearly 4 years for my tablet and 5 years for my phone.

And both battery`s seem to be still working fine.

John.

@NiMHi Those graphs don't seem consistent with the prior-reported numbers. The graph shows roughly 20% daily/session charges but the numbers imply 53%. Are you sure the numbers are correct?

Gauss163 said:

@NiMHi Those graphs don't seem consistent with the prior-reported numbers. The graph shows roughly 20% daily/session charges but the numbers imply 53%. Are you sure the numbers are correct?

Click to expand...

Yes, those are numbers the app shows me. Currently the app tells me that the estimated capacity (3,231 mAh) is based on 520 sessions with 25,818% charged for 834,212 mAh total.

Perhaps the app ignores the charging sessions with less than 20% charged. Or... I got the phone for 20 months which is around 600 days and I don't charge it everyday.... 520 sounds about right....

The second graph (blue/grey) shows the wear (cycles) for each day. Charging to higher percentage gives a higher wear number. So it's not % on the y-axis, but wear determined by the app.

NiMHi said:

The second graph (blue/grey) shows the wear (cycles) for each day. Charging to higher percentage gives a higher wear number. So it's not % on the y-axis, but wear determined by the app.

Click to expand...

I don't see how the app could determine wear from a single cycle. Only over time could it give an indication of real wear (decrease in capacity). And it has no way to determine what is actually causing that wear. It could be wear from the phone sitting in the sun in a hot car. It doesn't know.

All it's telling you is what its pre-programmed algorithm thinks should be causing wear.

WalkIntoTheLight said:

I don't see how the app could determine wear from a single cycle. Only over time could it give an indication of real wear (decrease in capacity). And it has no way to determine what is actually causing that wear. It could be wear from the phone sitting in the sun in a hot car. It doesn't know.

All it's telling you is what its pre-programmed algorithm thinks should be causing wear.

Click to expand...

Yes.

WalkIntoTheLight said:

I don't see how the app could determine wear from a single cycle [...]

Click to expand...

The bar graph is labelled "wear (cycles)", so presumably it plots the equivalent number of (full) cycles that you have put on the battery on each day, e.g. if you charged 30% in the morning and 20% at night then the total = 50% = 0.5 full cycles. But as I mentioned above that is far from consistent with the cumulative mAh it reports.

@NiMHi Is your daily cycle usage closer to 20% or 53%, i.e. do you use closer to 1/5 or 1/2 of capacity per day on average?

If the design capacity is 3000mAh but it estimates 3231mAh after 227 full cycles in 1.62 years then clearly it is way off.

The app might just be junk. That wouldn't be surprising. A lot of apps are banged out by some guy sitting in his basement, and has no in-depth knowledge of the subject. Since it's 100% reliant on the phone's internal battery sensors, which may be off quite a bit from the real battery discharge, it's using inaccurate data to try to predict something that is a far more complex multi-variate problem.

The iPhone 5 no longer receives OS updates, and the original iPhone no longer even has a compatible cell signal. Don't take something that's been made simple for you and complicate it..

Gauss163 said:

The bar graph is labelled "wear (cycles)", so presumably it plots the equivalent number of (full) cycles that you have put on the battery on each day, e.g. if you charged 30% in the morning and 20% at night then the total = 50% = 0.5 full cycles. But as I mentioned above that is far from consistent with the cumulative mAh it reports.

@NiMHi Is your daily cycle usage closer to 20% or 53%, i.e. do you use closer to 1/5 or 1/2 of capacity per day on average?

If the design capacity is 3000mAh but it estimates 3231mAh after 227 full cycles in 1.62 years then clearly it is way off.

Click to expand...

I think my daily usage is closer to 53%, but I'm not sure.

The calculated wear for each day is done differently. The higher you charge the battery the more wear. I don't know which algorithm they use. For example: The phone is now at 50% and charging to 80% (setting alarm at that %) will give according to the app 0.17 cycles in wear. Charging to 90% is 0.40 and to 100% is 0.92 in wear.

StarHalo said:

The iPhone 5 no longer receives OS updates, and the original iPhone no longer even has a compatible cell signal. Don't take something that's been made simple for you and complicate it..

Click to expand...

I'm doing it for the fun. Also, the Iphone 5 is from 2012/13 and I think 5 years is not asking too much from a phone, but the battery can make the phone unusable sooner. This is something the phone makers don't care about, because they can sell more if it is unusable after 2/3 years. But this is a bit off-topic.

NiMHi said:

The calculated wear for each day is done differently. The higher you charge the battery the more wear. I don't know which algorithm they use. For example: The phone is now at 50% and charging to 80% (setting alarm at that %) will give according to the app 0.17 cycles in wear. Charging to 90% is 0.40 and to 100% is 0.92 in wear.

Click to expand...

Ah, so that explains that apparent inconsistency. They are trying to weight the region of the cycles to reflect the higher degradation in extreme regions. Their docs don't say anything about that so it is not clear how they are doing it. You'd need massive datasets to derive algorithms that would work well for different chemistries, temperatures, etc. But that still doesn't explain the reason why it claims that your battery has 8% more than design capacity (3231 vs 3000mAh) after 227 equivalent full cycles in 1.6 years.