I'm having trouble clearing the BATLOWV bit in register 0x06 on the BQ25895. With VBUS connected, I am able to clear the bit. With VBUS disconnected and running only on battery power, the bit is stuck at 1. I can set all other settings just fine, including the rest of register 0x06. If I clear the bit with VBUS connected, it gets reset back to 1 when VBUS is removed.
Is this expected behavior? The datasheet as of Oct 2022 does not indicate any special conditions for access to the BATLOWV bit.
I've found this issue: https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/710331/bq25895-batfet-turn-off-at-vbat_otg-about-2-85v-in-boost-mode
The issue described has still not been documented (despite being 4 years old), that the BATLOWV setting is used to turn off BATFET in boost mode (instead of the battery depletion voltage in the specifications). This is unfortunately one of many instances of insufficient documentation for a very complex chip with a lot of hidden and unspecified behavior - is there any hope to ever getting the datasheet updated, or at least a detailed errata doc? Design errors happen, but we need to know about them in order to make our own designs work!
If this customer is actually getting a 2.85V cutoff, that would mean they have cleared BATLOWV somehow to enable the 2.8V setting on battery power, as the default 1 is 3.0V. The chip does seem to reliably cutoff at 3.0V in this case, so it is definitely using this bit. But what are the conditions necessary to clear the bit and get it to stay cleared?
Thanks!
