During UNDOT 38.3 Rev 7 testing of 16 assemblies of a reserve Li-Ion Battery pack containing the bq40Z50-R2, though the tests "pass" the criteria for the test, five of the units displayed the same failure mode in terms of operation. Using the schematic from the data sheet SLUSBS8B, page 28, Figure 21 as reference, the battery manager IC showed signs of thermal stress at pin 26 (VCC) and the resistor equivalent to R9 (100 ohm) was burned. The appearance is that of a free-flow of current through that resistor into pin 26 of the IC, possibly due to a latch-up within the IC, rendering it non-functional. The cells in the pack (3S1P configuration) were totally depleted. I suspect the test conditions for UNDOT38.3 may have exceeded one or more of the device absolute maximum ratings. Unfortunately we have been unable to determine the exact mechanism of the failure.
There is an odd configuration in the design to add redundancy that may have contributed to this situation. I would like to speak or chat privately with an application engineer to assist us in determining a root cause for this behavior and to evaluate the unusual design features employed to determine if they contributed.
