I setup and tested EVM boards to operate a Bq24630 DPM in a 24 V system using 6-cells of LIFEPO4 26650 cells (PCM included from suppilier). The output of the DPM drove an LM25118 buck boost converter EVM to operate a printer with 15A peak discharge currents. These EVM boards worked great over all my power supply and battery voltage requirements. Without getting into too much details, I put this system together as on a single board, following the PCB layout guidelines indicated in the data sheets,and artworks provided as best as possible, following the ground isolation recommendations. I 've built about 8 boards, but only 3 appear to operate as expected.
Debugiing is complex, and there are many questions concerning the BQ24630's erratic behavior, however, one general observation is that boards which do not operate correctly appear to have BATDRV~ or ACDRV~ ralied to ground or VCC (24V or the Battery Voltage ). The turn on voltage for the P-channels should only be VCC-6V, which is well within most P-channles Vgsmax of +-20V. When I remove all active parts from the board around the bq24360 there does not appear to be any shorts or netlist errors. If the Bq24630 is just removed the P-channels shut off thru the 100K pullups to the SYSTEM rail.
My question is, what external or internal events/conditions in the BQ24630 can cause the BATDRV~ or ACDRV~ P-channel gate drives to ground rail instead of going to Vcc-6V.
Thanks
Jim Del Signore