BQ76940: XREADY Fault on wakeup

Hi Vaibh,
CAP3 is the power supply for the top cell group circuitry which measures those cells. If the CAP3 voltage is incorrect it is unlikely the part can measure cell 15, or other cells in that group. With XREADY the cell voltage value may be residual. The voltage on cell 15 should not influence CAP3 other than the cell 15 voltage should be similar to the BAT voltage and the BAT - VC10X voltage minimum recommendation is 6V.

If the voltage from BAT to VC10X was 20V when the part was working and later was 10V with the same cell or source setup, the part is damaged and must be pulling significant current. For a 10V drop across a 1k Rf power filter resistor, current would be 10 mA instead of the typical 30 to 50 uA. If the current flows through other resistors to produce the low voltage the current could be different.

The board with cell 2 showing 200 mV between terminals with report of 65489 seems to indicate the inputs are shorted either externally or internally. The high voltage report would indicate damage to the part since 200 mV should be reported as 200 mV. A 200 mV value would be below the UVMINQUAL and would not report a UV. The OV is consistent with a high voltage. No XREADY is expected since the part is able to report the voltage.

The part was designed for random cell connection but does have abs max limits. For example some damage has been reported on cells 6 & 11 as shown in www.ti.com/.../slua749 figure 21, but this does not seem to be what you are describing.
Connecting one group at a time may prevent the possibility of pack voltages distributing improperly across the different groups. Connecting the bottom group, then the middle group then the top group should avoid excessive voltages. Booting after cell connection is recommended. Booting with only the bottom group connected will give an XREADY condition which should be able to be cleared once all groups are connected.

Hi Vaibh,
The TIDA-00449 and TIDA-00792 were both tested designs and should be good references. Sometimes things are drawn differently on schematics, so be sure all circuit features are included when assembling a schematic from different sources. A commonly lost section is the power supply filter components, but you have them on sheet 2.
Also on sheet 2, the bottom of C22 should be connected to GND as shown in the data sheet rather than VC0. The 2 TIDA circuits worked but it is recommended to follow the data sheet connection.
BQ76940 needs a REGSRC supply, it is cut off page 2.
BQ76940 also needs a boot signal to TS1, provide an appropriate signal for your system.
It is somewhat uneasy to proceed to a new design without having been able to resolve the XREADY issue with the present board. The TIDA references were successful and are trusted.
If you were using a MCU to clear the XREADY status, be sure you were writing "1" to the bit to clear after normal conditions were restored. Some ICs have a write 0 to clear and this is sometimes a confusion.

A supervisor such as TPS3839 is often used to reset the MCU when the operating voltage is outside the normal range, so holding the MCU reset until voltage is proper is a good idea for startup. There are system design considerations, note the features of your MCU. It could be beneficial for the MCU to know the supply voltage and take action rather than being reset. Consider if the MCU was balancing cells with the bq7640 and power went out of tolerance. If the MCU is held in reset the bq76940 would be left balancing and those cells would eventually deplete. If the out of tolerance voltage is a failure the battery has failed anyway, so it may be a suitable solution. An option may be to have the voltage monitor interrupt the processor to provide some shutdown sequence before stop.

Yes, the cells connect BAT- and the load/charger connects to PACK- in the diagram.