dumb and dangerous (?) phone repair

[olewales]

Hello, this is my first post on this forum so if I have done something wrong please don't kick me out right away. Feel free to shout at me though.
I just got myself an old motorola c121 mobile phone for playing with OsmocomBB project. I got it for 30PLN (which is about $10) as in "perfect working condition" with original battery and charger. Before potentially bricking it with osmocom firmware I threw in my backup SIM to see what is it like as a phone (it was fun experience to see stopwatch feature being translated into polish literally as "stop watch"). I quickly noticed that phone does not indicate charging the battery in any way (this may be by design, but I doubt it) and battery status indicator always shows full charge (when i was looking at it at least). Otherwise, it was working just fine. Later, when I was playing with flashing software I had some communication issues (which were my own fault btw) so I decided to tear down the phone to see if there are any obvious signs of damage.

What I saw disgusted me at first and after some more probing terrified me a little. There was obviously some failure in power/charging circuitry and someone attempted to repair it by shorting positive terminal of charging connector to positive terminal of the battery. I have verified that it is indeed a dead short. Now, this is why there is a question mark in the topic of thread. I am not a qualified EE but an unexperienced hobbyist. For me this is clearly not a way to treat a Li-ion battery but I am curious if it's as outrageous and dangerous bodge as I think.

Here are the facts:
- Charger has nominal charging voltage of 6.4V. Measured voltage is about 5.5V
- positive charging terminal is a dead short to positive li-ion battery terminal
- Negative battery terminal is NOT shorted to ground. There is some circuitry in between
- When charger is connected there is charger voltage (~5.3 under load) present at battery terminals at all times edit: probably not true and I screwed the measurement. It's difficult to get to the terminals with the battery in place. There is 5.3V on the terminals when the battery is not there though
- Phone works without battery inside when charger is connected and shows full charge on battery gauge edit2: battery gauge actually seems to be a software issue. ADC seems to work fine

I am curious about EEVblog community opinion on ingenuity of the guy who had done this. In fact the phone after this fix may be quite useful considering I managed to finally run osmocom bb stack on it and it works without battery inside.

[poorchava]

Lithium - ion cells in phones have protection circuits integrated into them.  They can be placed on positive or negative rail depending on design.  They typically disconnect the cell in case of overvoltage,  overcurrent,  over discharge etc.  So this may not necessarily be very dangerous,  but definitely will not make the cell happy either. 

[lapm]

Lithium - ion cells in phones have protection circuits integrated into them.  They can be placed on positive or negative rail depending on design.  They typically disconnect the cell in case of overvoltage,  overcurrent,  over discharge etc.  So this may not necessarily be very dangerous,  but definitely will not make the cell happy either.

Indeed they have if they are good brand and not cheap Chinese knock-offs.. There was recently case in my country where official manufacturers supply chain had managed to deliver one of these "fake batteries".. Consumer officials are all over the case... Since that battery happened to decide to turn into firebomb... No protection circuit, no over presure valve, etc..

[olewales]

Indeed they have if they are good brand and not cheap Chinese knock-offs..

That's what I initially thought of when I opened it up. Using included battery I am probably safe since previous owner obviously used it in this state and it did not burn down. I am surprised that the Li-ion cell still works and charges! Now, when I think about it I am pretty sure I screwed something up during voltage measurement. There must be still some kind of regulator on the negative rail because otherwise the protection circuit should cutoff immediately after the battery receives over 5V on its terminals. The spec (and by spec I mean wikipedia article) says that protection usually kicks in above 4.3V).
Maybe it's only monitoring circuit that was bypassed and charging controller is independent of it.

OR mayble I am just paranoid, this is just valid fix for a broken trace and a device is working just fine however those are things why I think that something was bypassed in the process:

- Phone works fine without the battery. For me this is strange because I never saw one that does. Even most modern devices seem to use a battery as some sort of "buffer" and will refuse to start (on AC charger) even if the battery is connected but empty.

- No indication of charging process and "full battery" indicator 100% of the time. I will verify this by discharging the cell and I will watch the indicator carefully.

I am now genuinely curious about the nature and the cause for this repair. I'll probably eventually buy identical device just to verify how it works.

[amyk]

This reminds me of the little "MP4 players" which were popular a few years ago --- 4.2V lipo with built-in protection, connected to Vbus through usually nothing more than a resistor and diode. Here's an example:



They definitely did not have any sort of charge controller; the battery level was measured through a divider and ADC, so you would see "full battery" every time it was plugged in and not charging; the on/off switch just connected/disconnected the battery from Vbus. The fact that numerous reports of them catching fire and so forth have not appeared suggests that it wasn't as dangerous as it would seem.

[olewales]

Ok, so I want to sum it all up. I did some additional research and the phone is probably not as broken as I initially thought. I am posting this just in case someone finds (by the miracle of google probably) something useful in it.

Battery gauge meter actually does work, but it seems to sample battery voltage very rarely. During my discharge test the phone actually shut down itself (not "gracefully", it seemed more like some brownout detection kicking in) and after powering it up again (without charging it first) it actually showed "Low battery" warning on the screen.

Additionally I noticed that OsmocomBB layer1 firmware prints some power status data on the console.

Here is the sample of what comes out without the charger connected:

Code: [Select]

BAT-ADC: 556   1   0   0 1023 418 372 216 
Charger at 9 mV.
Battery at 3801 mV.
Charging at 0 mA.
Battery capacity is 75%.
Battery range is 3199..3999 mV.
Battery full at 468 LSB .. full at 585 LSB
Charging at 239 LSB (204 mA).
BCICTL2=0x3ff
battery-info.flags=0x00000000
bat_compal_e88_chg_state=0
And while charging (right after connecting DC power plug):

Code: [Select]

BAT-ADC: 593   1   0   0 1023 519 457 236 
Charger at 9 mV.
Battery at 4054 mV.
Charging at 0 mA.
Battery capacity is 100%.
Battery range is 3199..3999 mV.
Battery full at 468 LSB .. full at 585 LSB
Charging at 239 LSB (204 mA).
BCICTL2=0x3ff
battery-info.flags=0x00000000
bat_compal_e88_chg_state=0
This clearly shows that hardware inside is still capable of measuring battery voltage, but seems to also confirm that the charging monitoring part of the circuit is broken (I assume that log lines that mention charging current should normaly state something meaningful). However the charging seems to work just fine. Battery seems to be in very good condition.

I will leave it there for now. Thank you to all that participated in this topic. The design of the circuitry remains mystery to me but I learned a few new things while investigating this problem and this matters to me.