Samsung phones are charging batteries with an ugly pattern?

[jjman]

Hi Guys,

Looks like there is a very ugly charging pattern on all samsung galaxy phones I could get hold of. Can you please check these results and tell me if there is some obvious problem with my conclusions?

So here is the graph (see the attachments):
time: The time since the first measurement interval
agilent current: is a charging current measured on the USB port using Agilent U1272A + U1177A
phone current: is the charging current reported by the current sensor on the phone

From the chart it looks like when the battery is fully charged instead of detaching it and running straight on power from the AC (like a normal notebook) the phone starts running off the battery then discharges it a bit (the battery gauge on the phone shows always 100%) and then recharges it again. So at the end it looks like the battery management system is doing microcycles right in one of worst areas possible: the 90-100% capacity area.That should promote the forming of the solid electrolyte interphase hense speeding up battery degradation (as we essentially have the battery operated on the top voltage which increases the oxidation of  cathode from the electrolyte and it makes the SEI layer thicker) .

Now why is that a problem? Well … when usually most people charge their phones? During the night … hence looks like a very nice planned obsolescence ‘feature’?

P.s.
Here is my setup if you would like to reproduce it:
1. Pc is pulling data from agilent via BT with SPP profile (essentially using the FETC? command)
2. Pc is pulling data from app on the samsung (that is again exposed with SPP and responds to the same FETC? command). The app uses the /sys/class/power_supply/battery/batt_current_ua_now kernel driver to read the current sensor from the BMS.
3. There is some slight desync (in the ms range) as both sources of information does not respond simultaneously.

CSV files with the results are attached here.

Best Regards
Ivan

[amyk]

Have you tried phones of any other manufacturer?

I suspect this could be due to a "serial" power management system (charger<>battery<>discharger) and done to save cost...

[jjman]

It looks like it is not cost related, it looks like deliberate battery degradation ... and now that you mentioned cost here is example from a cheap Xiaomi (redmi note 2).

Once charged it does not use the battery.

My interpretation of the results:
As the phone is cheap it does not have a current sensor inside
(do not be fooled by the data in the /sys/class/power_supply/battery/curent_now kernel driver - it is 0 for discahrging, 515 on usb charger and 1468 for AC charger).
So one can see that once charged it continues working in the 200mA-500mA current draw.
When video is played (to check the charging behavior) once can see that the phone just increases the consumption from the charger (a solid 600mA) once the video playback finishes the current consumption goes to the old state which is observed when the phone is in standby and fully charged. The important thing is that during the video playback no "charging" pattern is seen. And when the phone is fully charged and in standby current continuously flows in to the device which for me is an indication that the battery is not used.

Attached are some sample files.

[Mechatrommer]

It looks like it is not cost related, it looks like deliberate battery degradation

really? any reference to this claim? one of our galaxy's (kzoom) battery hasnt been changed since longer than 5 years now iirc. how long should we expect a lion battery lasts on constant use? 10 years? 20 years? and what profit will they make if we buy 3rd party battery as replacement?

[Yansi]

Complete stupid and dumb ignorance, if not anything else.

[mairo]

Could it be something related with the recent battery issue that they had with one of theirs new phones (do not remember the model number, 8, or 9)? On how many different types of Samsung phones did you test it ?

[amyk]

It looks like it is not cost related, it looks like deliberate battery degradation ... and now that you mentioned cost here is example from a cheap Xiaomi (redmi note 2).

Once charged it does not use the battery.

You've discovered the difference between what the industry calls a "battery fed" design vs a "charger fed" design. Here's a good explanation of the basics:

https://www.analog.com/media/en/reference-design-documentation/design-notes/dn336f.pdf

With the "series" or "battery fed" design (which the Samsungs seem to have), the load supply is taken from the battery, so in your use-case the charger detects the battery is not full and begins charging, then shuts off once the battery is full --- allowing the load to discharge the battery below the "begin charging" point, and the cycle repeats. This is a cheaper topology since there's no real switching of power taking place; just one charger.

With the "parallel" or "charger fed" design,  the load is connected directly to the power input and the battery through a charger/discharger. The power input both charges the battery and powers the load. Once the battery is charged and VBUS > VBAT the charger turns off and the load is directly powered from the input. The Xiaomi you have uses the common MTK platform, and (I also have an MTK phone, not a Xiaomi however, so can confirm) the PMIC on those appears to be based on the charger-fed topology.

[jjman]

really? any reference to this claim?

There are graphs and data and attempt to analyze in this very topic ... would be such a joy if there was someone to read them

[jjman]

You've discovered the difference between what the industry calls a "battery fed" design vs a "charger fed" design. Here's a good explanation of the basics:

https://www.analog.com/media/en/reference-design-documentation/design-notes/dn336f.pdf

Thank you! The DN was a very informative read
Now it makes way more sense. It looks like they use battery fed design for their flagship phones.

[jjman]

I have tested

Could it be something related with the recent battery issue that they had with one of theirs new phones (do not remember the model number, 8, or 9)? On how many different types of Samsung phones did you test it ?

I have tested s2 s4 and j5 they all show the same pattern. Check the answer from @amyk it gives a very good explanation of the observed results.

[Mechatrommer]

really? any reference to this claim?

There are graphs and data and attempt to analyze in this very topic ... would be such a joy if there was someone to read them

saw that. just because of that kind of shape, do you have right to say its a degradation attempt? is that a thesis or what? we need better proof, such as statistical data of batteries life vs charge/discharge cycle similar to what you have shown. do you have piles of dead batteries already from exhaustive test? if not, i can also say, xiamoi attempts to degrade charger, or toyota degrades their cars because they let them move, or God (or nature or whatever "something" (look at my signature) you like) deliberately degrade creation by letting them die old or mutilated born or whatever. there are always something to blame, without giving undisputable proofs and deeper level of thought

[timelessbeing]

Yeah. Calling it an inferior design and profit maximizing would be fair. Sabotage? You better have good proof, otherwise it's libel.

[jjman]

we need better proof, such as statistical data of batteries life vs charge/discharge cycle similar to what you have shown. do you have piles of dead batteries already from exhaustive test? if not, i can also say, xiamoi attempts to degrade charger, or toyota degrades their cars because they let them move, or God (or nature or whatever "something" (look at my signature) you like) deliberately degrade creation by letting them die old or mutilated born or whatever. there are always something to blame, without giving undisputable proofs and deeper level of thought

If you want to contribute to the discussion please share how my interpretation of the data is wrong. E.g. how the current charging behavior is NOT leading to battery degradation. Do you have something to say?

p.s.
Here is why I think it is bad (sorry, requires reading: https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries)

[jjman]

Yeah. Calling it an inferior design and profit maximizing would be fair. Sabotage? You better have good proof, otherwise it's libel.

Year, note taken, bad wording choice from my side (English is not my native language). I though that saying "it looks like" gives enough shakiness to the statement.
All what I wanted to say is "Looks like this promotes battery degradation" (if it is on purpose or not is a whole different discussion that has nothing to do with electronics).

[The Soulman]

You've discovered the difference between what the industry calls a "battery fed" design vs a "charger fed" design. Here's a good explanation of the basics:

https://www.analog.com/media/en/reference-design-documentation/design-notes/dn336f.pdf

Thank you! The DN was a very informative read
Now it makes way more sense. It looks like they use battery fed design for their flagship phones.

It is also possible that a higher peak current is needed than a usb charger could reliably deliver, perhaps for a tx burst. 

[aheid]

It is also possible that a higher peak current is needed than a usb charger could reliably deliver, perhaps for a tx burst. 

When using my S5 for navigation during bike trips in the summer, my external battery pack can't keep up and the onboard battery discharges, though a lot slower than if the external battery isn't connected.

That's with backlight to max due to sunlight, rendering of map, data communication for loading map, and of course GPS. In addition some ANT+ sensors to track.

[Mechatrommer]

we need better proof, such as statistical data of batteries life vs charge/discharge cycle similar to what you have shown. do you have piles of dead batteries already from exhaustive test? if not, i can also say, xiamoi attempts to degrade charger, or toyota degrades their cars because they let them move, or God (or nature or whatever "something" (look at my signature) you like) deliberately degrade creation by letting them die old or mutilated born or whatever. there are always something to blame, without giving undisputable proofs and deeper level of thought

If you want to contribute to the discussion please share how my interpretation of the data is wrong. E.g. how the current charging behavior is NOT leading to battery degradation. Do you have something to say?

p.s.
Here is why I think it is bad (sorry, requires reading: https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries)

because you made a claim, i didnt, so you have to provide the references. i didnt say you are wrong, i only asked for references. except me from my observation that galaxy's battery can last up to more than 5 years which contradicts from your single case statistical plot, but that only 1 sample of statistical study which is poor on both of us, for me at least i know its poor. and you still dont give us good answer on what benefit can samsung gain by deliberately try to degrade battery life? and why? you need also to give your logical reasoning if you want to make a claim. from your link you still dont have a strong standpoint to make such a claim and i believe you have some misunderstanding there. for example number of cycle stated is cycle of full to flat usage. your data is similar to trickle charging (trickle on fully charged battery), which is recommended by most if not all battery manufacturers/ researchers. sorry if i have to ask you to read about trickle charging.

and even the battery manufacturer/researcher is unable to make a conclusive statement on this... from your link...

Evaluating battery life on counting cycles is not conclusive because a discharge may vary in depth and there are no clearly defined standards of what constitutes a cycle (see BU-501: Basics About Discharging)

not as bold as, errr....

btw, this one needs reading as well...
https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

[NiHaoMike]

If you really want to boost battery lifetime, stop charging at around the 80-90% range. Could be done by reprogramming a smart plug but one of those push button timers is a ready made solution that's "close enough".

[jjman]

btw, this one needs reading as well...
https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

Yes!

Choosing a lower voltage threshold or eliminating the saturation charge altogether, prolongs battery life but this reduces the runtime. Chargers for consumer products go for maximum capacity and cannot be adjusted; extended service life is perceived less important.

Do you agree keeping the battery at max voltage is bad? If you say "no definite answer can be given" than what experiment would you like to see to have enough evidence? (I have some ideas but do not want to spoil the thread)

[jjman]

It is also possible that a higher peak current is needed than a usb charger could reliably deliver, perhaps for a tx burst. 

Yep, that is also a good explanation (the data is based on 2 sec measurement interval no idea what happens in between), but ... judging from the link posted by @aymk it is not probably the case.

[ogden]

So at the end it looks like the battery management system is doing microcycles right in one of worst areas possible: the 90-100% capacity area.

When during day you charge your phone, leave it for let's say 3 hours what battery charge state you expect it to be when you grab it to go out and do whatever you want to do? 89% or 100%?