Internally the pictures I've seen of these Chinese batteries all have a variation on the same logic board mine has. There is an MCU, a protection chip, and a voltage regulator. These boards have 3 labeled pads marked A, B, and C. These tell the MCU what capacity cell it is connected to. There is also a set of 5 pads, which in my version were labeled G,R,V,C,D. G is ground (B-), and R is connected directly to the reset pin of the MCU. I shorted R and G momentarily, which rebooted the MCU and then I reassembled the battery. Now its back in service.
I have seen others suggest draining the battery with a diode, but the MCU can operate all the way down to 2.7v so this would be bad for the cell. These cells are of dubious quality to begin with. I have also seen someone suggest unsoldering the cell, which would reboot the MCU, but with the reset pin so nicely exposed that seems like overkill.
While I was looking over the board, I see why these are so problematic. There is no fuel gauge chip or coulomb counter. The only indication it has of the state of charge is the delta between the regulated voltage and the battery voltage. There is no eeprom to store charge cycle counts, or other cell related data and the cell capacity is set by shorting I/O port pins to ground. The software does not appear to implement enough of the communications protocol to charge from a PSP-330 battery charger, or set/get the serial number. The MCU is active at all times, even if the protection chip were tripped by an undervoltage condition.
On the plus side, it may have some hacking potential. The G R V C D test points appear to support in-place programming of the MCU. It may be possible to tinker with the code.
----
A little update. Since I posted this I have had to reboot all of my clone batteries with the exception of the lenmar. They seem particularly prone to crashing during USB access/charging.
Since I have had to open them all up I checked the cells. My "1200mah" clones have unmarked 063450A cells which puts them at ~1000mah. My "3600mah" clones have a mix of (marked) 063450AR and 063450ARH cells which mean the best of the lot has maybe 1100mah of actual capacity. Only one of the cells carries the UL registered component mark. By comparison, an OEM 1200mah battery has a US633450 A9H with a design capacity of 1250mah.
In short, they all suck. So I am harvesting the best cells of the lot to repair a couple of OEM slim batteries that died from bloat. Sounds simple enough but the eeprom data needs more study.
Advertising