Li-Ion charging strategy

Status
Not open for further replies.

etc

Flashlight Enthusiast
Joined
Dec 19, 2004
Messages
5,777
Location
Northern Virginia
So I have a samsung device, s8+, that has an internal 3500 mAh cell.
I installed the phone inside an external ZeroLemon Li-ion battery that has 8500mAh capacity.

Now the external battery only charges the primary when it is turned on. That raises the question, when should the extended battery be turned on? Always? Or when the primary drops below say 4v? Or at 3.7v? There is a physical button you have to push.
 

archimedes

Flashaholic
Joined
Nov 12, 2010
Messages
15,780
Location
CONUS, top left
What is your primary purpose ?

To maximize your continuous runtime, or to maximize the duty life of your non-replaceable internal battery ?
 

etc

Flashlight Enthusiast
Joined
Dec 19, 2004
Messages
5,777
Location
Northern Virginia
To get the most years out of both internal and external Li-ion cells.

My cylindrical 18650s I recharge at about 3.9v +/-
 

archimedes

Flashaholic
Joined
Nov 12, 2010
Messages
15,780
Location
CONUS, top left
The optimal strategy is challenging, since two of the largest negative factors on rechargeable cells work at cross-purposes ....

• time spent at charge extremes (especially at > 80-90% SoC, and most especially when also at relatively high temperature)

• repeated duty cycling

So, keeping SoC between (say) 20% - 80% will reduce loss of capacity over time, at the expense of increased cycling

Or conversely, minimizing the total number discharge / charge cycles ( by charging to 100% and draining to 0% ) will unfortunately stress the cell through increased temperature and aging effects.

The theoretical "ideal" strategy may be somewhere between those extremes, but I tend to favor avoiding extremes of SoC, in the case of "general mobile phone" type usage
 

etc

Flashlight Enthusiast
Joined
Dec 19, 2004
Messages
5,777
Location
Northern Virginia
The axioms I learned are:
Overcharging is bad.
Over discharging is bad.
Overheating is bad.
The more cycles, the worse lifetime becomes.

I posed the very same question almost 10 years ago when I first got into Li-ion and I dont remember the consensus if a partial cycle is as bad as a full cycle.

Given the capacity ratio is about 2.4:1, I can deplete and recharge the internal battery 2.4 times. However this is very slightly different from 18mm cells because the charger, if you will, is always plugged in.
 

InvisibleFrodo

Enlightened
Joined
Sep 16, 2014
Messages
963
There is a website called battery university that has a lot of good information and a lot of data from real world tests. One thing I found there is that as a general rule, every time peak voltage comes down 0.1 volts, the expected cycle life of the cell will double. Meaning if you were charging your Li-ions up to 4.2 volts per cell and you could expect about 500 cycles, you should theoretically be able to cycle the same cell 1000 times if you charged the cells to 4.1 volts. The reason is the decrease in voltage stress that cells go through when they are at relatively high voltages. NASA charges the cells in their satellites to 3.92 volts per cell because they say that eliminates all voltage stress.

I'm having a harder time finding out what is the ideal voltage to bring the cells down to. I have been generally charging when the cells are down to 3.8 volts or 3.7 volts. Call me crazy, but since the nominal rating on the cells is 3.7 volts, coming down to that state of charge shouldn't be harmful in my opinion.

Anybody know about what is the ideal voltage to discharge down to?
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
There is a lot of controversy about battery university. Some of their articles is more or less copied from other sources without them verifying the results.
The general assumptions is that lifetime drops when LiIon batteries are charged to a high voltage and there is some evidence to support it, but how much for any of the never chemistries is not really verified.
I will probably never test that, I do not like to do multi years test, anything that takes more than a month testing time is problematic for me.
 

iamlucky13

Flashlight Enthusiast
Joined
Oct 11, 2016
Messages
1,139
I'm seeing two options. As you probably know, lithium-ion batteries experience more aging when stored at full charge compared to when stored at partial charge, but cycling them also wears them out. I don't know which would actually be better:

(A) Leaving the external battery turned on all the time so the phone battery stays fully charged, meaning it will experience more aging but less cycling wear.

(B) Trying to follow a specific cycling strategy, meaning less aging but more cycling wear.

If you are using the phone really heavily, meaning you'd accrue a lot of cycles over a given time, my gut is to go with A. If you'd normally get over a day between charges, I might be tempted to try B.

If you choose a cycling strategy, generally, if you can keep the battery voltage in the middle of its normal range, you should see the best cycle life - so perhaps start at ~3.3V and charge to to ~3.9V.

Gauss163 has shared a paper a few times that shows a handful of tests regarding this. If I remember right, compared to cycling batteries from 0-100% state of charge, or even 50-100% SoC, they saw a significant increase in total energy delivered (not just cycles) over the life of the cell by cycling from 25-75% SoC. I've seen other sources recommend a 20-80% cycling range. I don't know if this is related, but I think 75-80% charge is usually right at the edge of where charging switches from constant current to constant voltage. It wouldn't surprise me if cycling wear increase somewhat at that point.

Can you reliably control the charge this way, however? I guess since smartphones have their own battery meter that I think works mainly by Coulomb counting, you hypothetically could know SoC with decent accuracy directly, instead of relying on voltage as a proxy for SoC. Every now and then (maybe 100 cycles or so?), you'd probably want to do a full discharge until you get a low battery shutdown warning and then a full charge in order to recalibrate the meter against drift and capacity loss. I'm assuming low voltage shutdown on the S8 is based on voltage.

Is it worth the effort of the extra attention required, though? I suppose it's just something you could try to get in the habit of doing when you think about it, and not worry too much if you forget, because at least for those times you remember, you're managing the battery life better than most people.

It would be harder to control the cycling of the external battery, but even if you don't get comparable life out of it as out of your phone, it's at least less expensive to replace, or you just keep using the phone without it.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
A few posts above contain remarks that might prove misleading. Generally cell life increases if we use shallower centered cycles, because this minimizes the time that the cell spends at extreme voltages (which greatly accelerate major degradation processes).

Perhaps a specific example will help to clarify. Let's suppose that we only do 50% depth cycles, and we center them halfway, i.e. we use only the capacity range from 25%-75%, i.e. we only charge to 75% capacity, and we only discharge to 25%. One study shows that this doubles the life of the cell, i.e. we'll get twice as many total Ah (= cumulative capacity) over the entire life of the cell. For details see the Wöhler Curves I posted here.

The tradeoff is that because each partial cycle (25%-75%) yields only half the capacity of a standard full cycle (0%-100%), we'll have to charge our device twice as frequently as we would if we used full cycles. This is not an issue if charging it automated, but could prove too inconvenient for manual charging.

Using even shallower depth discharges can yield even greater gains in life. As an extreme example, the above-linked study shows that we can get 17 times the life by using extremely shallow cycles (the middle 5%).

Regarding the question in the OP: try to keep both the internal and external batteries in the range 10%-90% if possible (or, better, 15%-85% if that is convenient). There are apps (e.g. GSam Battery Monitor) that you can use to signal alarms at upper and lower limits. Most important is to never let the battery remain near 100% for long periods. If you accidentally charge to 100% then immediately run some high-drain apps to get it down to at least 95% asap. Charging just before heavy periods of use will minimize the time that the battery spends at high voltages, so will help to maximize life, e.g. charge in the morning (not overnight - which leaves the cell at high voltages all night).
 
Last edited:

etc

Flashlight Enthusiast
Joined
Dec 19, 2004
Messages
5,777
Location
Northern Virginia
There is a website called battery university that has a lot of good information and a lot of data from real world tests. One thing I found there is that as a general rule, every time peak voltage comes down 0.1 volts, the expected cycle life of the cell will double. Meaning if you were charging your Li-ions up to 4.2 volts per cell and you could expect about 500 cycles, you should theoretically be able to cycle the same cell 1000 times if you charged the cells to 4.1 volts. The reason is the decrease in voltage stress that cells go through when they are at relatively high voltages. NASA charges the cells in their satellites to 3.92 volts per cell because they say that eliminates all voltage stress.

I'm having a harder time finding out what is the ideal voltage to bring the cells down to. I have been generally charging when the cells are down to 3.8 volts or 3.7 volts. Call me crazy, but since the nominal rating on the cells is 3.7 volts, coming down to that state of charge shouldn't be harmful in my opinion.

Anybody know about what is the ideal voltage to discharge down to?

That's really interesting. Thanks for sharing. I read the battery university site some years ago but had forgotten most of what I read -need to visit it again. It is a great, informative sight.
 

InvisibleFrodo

Enlightened
Joined
Sep 16, 2014
Messages
963
There is a lot of controversy about battery university. Some of their articles is more or less copied from other sources without them verifying the results.
The general assumptions is that lifetime drops when LiIon batteries are charged to a high voltage and there is some evidence to support it, but how much for any of the never chemistries is not really verified.
I will probably never test that, I do not like to do multi years test, anything that takes more than a month testing time is problematic for me.

I saw a symposium on Li-ion degradation and the causes of cell death. I don't remember the university running the tests, but they covered the internal workings along with an entire list of different additives that can modify cell behavior. The best additives would help prevent excessive buildup on the cathode if I'm remembering this all correctly, and that allowed the cells to cycle many more times. They also tested groups of batteries being charged to different maximum voltages. The voltages were 4.25, 4.35, and 4.45 volts with very obvious decreases in cycle life as the voltages went up.

There is really real rea search out there that really says higher voltages will decrease cycle life.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
[...] I read the battery university site some years ago but had forgotten most of what I read [...]

That's good that you forgot most of it because Battery "University" is not a reliable source for accurate information. You won't find any experts who recommend that site. It is blacklisted by Wikipedia.
 

InvisibleFrodo

Enlightened
Joined
Sep 16, 2014
Messages
963
That's good that you forgot most of it because Battery "University" is not a reliable source for accurate information. You won't find any experts who recommend that site. It is blacklisted by Wikipedia.

It's weird that Muyshondt has a link to battery university on his website and recommended it as a good source of information... [emoji102]🤔
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
It's weird that Muyshondt has a link to battery university on his website and recommended it as a good source of information... [emoji102]樂

I'm not aware of any Li-ion consumer-level flashlight and charger manufacturers that employ Li-ion battery experts. They are typically found only in large top-tier companies and research universities. (e.g. Panasonic, Tesla, MIT, etc).
 

InvisibleFrodo

Enlightened
Joined
Sep 16, 2014
Messages
963
I'm not aware of any Li-ion consumer-level flashlight and charger manufacturers that employ Li-ion battery experts. They are typically found only in large top-tier companies and research universities. (e.g. Panasonic, Tesla, MIT, etc).

So then who did he hire to manufacture his 18350 and 18650 cells?
So battery, flashlight, and charger manufacturers never employ battery experts? Who's designing these cells then? Who's designing these chargers?

I guess I don't need to ask where all the battery experts are, because you already told me. Thank you.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
So then who did he hire to manufacture his 18350 and 18650 cells? So battery, flashlight, and charger manufacturers never employ battery experts? Who's designing these cells then? Who's designing these chargers?

It requires very little Li-ion battery expertise to design (simple) Li-ion devices such as flashlights and chargers, since there are widely available off-the-shelf components available that have been designed by experts (charging ICs, BMS, etc).

But it does require significant expertise to write an encyclopedia covering all aspects of (Li-ion) battery electrochemistry - knowledge of which the author of BU = Battery "University" greatly lacks.

In case it is not obvious, it is worth emphasis that BU is essentially a giant infomercial for Buchmann's battery charger company Cadex (there are over 400 mentions of "Cadex" in this "university"). The primary purpose of the site seems to be to boost the reputation of the author and his company - not to provide accurate scientifically vetted information on battery technology. It is one of the most glaring examples of pseudoscience in the history of the internet. That's why Wikipedia does not allow you to link to it. The information on BU is no more reliable than the information on a late-night TV informercial.
 
Last edited:

etc

Flashlight Enthusiast
Joined
Dec 19, 2004
Messages
5,777
Location
Northern Virginia
I've had a lot of experience with ZeroLemon Li-Ion cells. They are much like 18650s except flat and rectangular of course. Over the years, I've had about 5-6 of them and used them heavily over these years. It seems that loading them to full capacity every day does diminish their ability to take full charge in about 6 months.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
[...] It seems that loading them to full capacity every day does diminish their ability to take full charge in about 6 months.

What do you mean by "loading them to full capacity"? If you mean keeping them fully charged then that is one of the worst things you can do to a Li-ion battery since such high-voltages greatly accelerate major degradation processes.
 

InvisibleFrodo

Enlightened
Joined
Sep 16, 2014
Messages
963
What do you mean by "loading them to full capacity"? If you mean keeping them fully charged then that is one of the worst things you can do to a Li-ion battery since such high-voltages greatly accelerate major degradation processes.

Not that I'm disagreeing with you, because I know what you just said is right, but I'm just wondering how you know for sure. Where has this information come from? Is it anecdotal, or are you quoting one of your battery experts? Is there a particular study you are referring to?
 
Status
Not open for further replies.
Top