The board has IC DW01. And the circuit on the board is the same as the datasheet describes.
Interestingly I ended up making a custom board, and I ended up connecting pin 6 of IC DW01 to the GND of the load circuit instead of connecting pin 6 of DW01 to the battery negative as the datasheet says.
I really don't know if the manufacturer made a mistake in the datasheet circuit,
But when I only tested the protection circuit as suggested by the DW01 datasheet (DW01 + MOSFET's), without the charging part of the TP4056, when reversing the battery polarity, there was a current of approximately 30mA.
And the voltage on the "BATT output" pins appeared inverted, without any voltage drop, that is, the circuit allowed 100% of the inverted voltage to pass (4V).
On the other hand, with the modified circuit, connecting pin 6 of the DW01 to the pin described in the datasheet circuit as "BATT-" (board output, not battery negative): there was no current when inverting the battery, and the "BATT output" there was no electrical voltage (0V).
If anyone can test this DW01 better please comment the results.
I think it would be good to fix the schematic of all the boards that have already been installed and I believe there are thousands of these boards in use.
You will find when the OC or OD circuit activates you will loose the negative battery connection to the chip. A quick guess would be it will oscillate and slowly continue to over charge or over discharge.
Hi, yes you are correct, about the loss of the battery negative pole reference when the DW01 protection activates.
I did some tests and got to this scheme, I added a low power MOSFET to ground the DW01's GND only if the polarity is correct.
Apparently it worked, there was no electrical voltage at the output when the battery was inverted, and apparently there was no significant electrical current as there was before the MOSFET was added:
If this is enough to protect against polarity reversal, I wonder why this hasn't been built into the DW01 IC.
Perhaps the DW01 was just designed to be integrated into the battery rather than the consumer device.
In this case, it was an error to have used this IC DW01 in this type of board with IC TP4056, which will be incorporated in the consumer device.
The developer cannot use an unprotected removable battery. The developer will have to use special connectors that don't allow polarity inversion, because some users have a heavy hand and make the connector go in backwards.
The datasheet already says this (and now I see that the limitation is a way of saying that the circuit doesn't have reverse battery polarity protection):
General Description
The ultra-small package and less required external components make it ideal to integrate the DW01-P into the limited space of battery pack
Applications
Protection IC for One-Cell Lithium-Ion / Lithium-Polymer Battery Pack
These are just hobby boards so user beware. The board I have has an additional function being a 3.7 to 5.0 up converter. When I looked at the "typical" schematics for each device the designer of my board simply stacked the three together, absolutely no schematic changes.
I may be wrong but I see the 18650 comes with a typical positive bump (like any AA battery) the battery holder can be designed to eliminate connections if the battery is installed reversed.
I tried to reproduce the circuit with MOSFET but it didn't work, so I noticed that the pins 1 and 3 of the MOSFET were reversed, for some reason when I saw some reference on the internet I ended up using the MOSFET in reverse.
I also tried using a P-channel MOSFET to switch VCC instead of GND, but it didn't work, about -2.5V came out.
The MOSFET's internal diode cannot allow the DW01 to flow current when the battery is in reverse polarity. The scheme that worked is this:
I have a small UPS with one 18650 cell. It has been working for about a year. A few days ago the UPS shutdown without notice. Output was zero based on my Hubitat Hub having no LEDs illuminated.
Troubleshooting found the Li-Ion cell to be an open circuit. This caused the overvoltage to shut down the device.
Trying to determine why the Li-Ion cell failed led me to find the life of a Li-Ion cell drops significantly when "floated" above 3.8Volts.
So now I'm trying to determine how I can loose 1/2 a volt on the battery. I am looking at Schottky diodes in series with the TP4056 or perhaps back to back in series with the battery.
I've had problems with emergency lighting. Even with a very low charging current, around 25mA, the luminaire had no protection to prevent overvoltage in the batteries.
I had to install boards with TP4056 on them, but I changed the current adjustment resistor so that the TP4056 didn't heat up too much and caused bad contact in the soldering.
After modifying it, the batteries have been working without any defects for years.
