BQ77910A and 2-wire solution

Hi Pierrick,

The bq77910A will have the best selection of features if the CHGST can be controlled by a signal which indicates the charger is attached.  Most circuits, particularly simple circuits will compromise some of the flexibility of the feature selection.  With a latched 2 wire circuit where the pack has only a PACK+ and PACK- terminal and CHGST is derived from a simple 3 transistor circuit, the pack can be shut down from TS and awakened by charger voltage.  It then keeps CHGST high until a fault occurs which turns off CHG. So the part operates most of the time as if it is 'in the charger'.  A consequence is that CHGST controlled balancing modes are not effective, and mode 1 is the only reasonable.  The advantage over CHGST tied high is that SCC faults can be recovered.  SCC can't be turned off, it can be set high, but if it were ever to occur when CHGST is permanently high, it can't be recovered since the recovery mechanism is to take CHGST low. Similar to CHGST tied high, a system with the latched 2 wire circuit enters UV with CHGST high, so CHG remains high and the part does not shut down from the under voltage condition.

The 2 wire circuit without the latch is still possible with the 910A.  It has limitations.  Since CHGST will only pulse high until the part wakes up and turns on CHG, again balance modes 2 and 3 are not usable.  SCC faults may cycle the CHG FET since the recovery condition of CHGST low is met at the time the fault occurs.  The part will shut down with under voltage, but when the charger is connected it will wake up and turn on the CHG FET.  The part may cycle in and out of shutdown pulsing the CHG FET until the cell voltages incrementally increase to come above the under voltage hysteresis threshold.

The rev A version eliminated a shutdown-with-regulator-on option, it did not change the UV fault recovery. The 2-wire without latch method will still wake the part but can cycle as described above, so the latched circuit was chosen as the common suggestion for 2 wire packs. 

Hi WM5295,

Thank you for your answers. I think i have almost all the answers i a need, but there is still something that i want to be totally sure.

I have a documentation from TI called "bq77910 & bq77908 Update CHGST Glitch Issue and Fix". In this document, it is explained that an issue can occur if there is an UV fault with a shutdown, SHTDIS =1 (LDO = ON) and the 2-wire solution implemented. And, in this document, TI asked everybody to put SHTDIS =0 (Rev.A put this bit to 0 in the default settings) and pleased everybody to use the improved “latched” 2-Wire circuit".

So, if i use the 2-wire (without latch) circuit with a bq77910A ( RSVD1/SHTDIS =0) and UV_REC_DLY=1, the issue of non OV cells protection after an UV shutdown will NEVER EVER happen, is that right?,

Thank you, best regards.

Pierrick.

Thank you very much.

Best regards,

Pierrick.