• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 295 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ24193: Risk with not placing the reserved bit REG07[3] properly + contradication in the datasheet

THOMAS PORCHEZ
Prodigy 50 points
Part Number: BQ24193

Hello,

          In a battery powered device, we use the BQ24193 component in combination with a single cell battery. In this design, we sometimes (quite rarely actually) encounter a situation where we want to go into shipping mode (section 8.3.1.2.2 of the datasheet), and it seems that the battery remains activated, and thus our system doesn't power off.

During the power off sequence, our microcontroller sends the power off commands to the BQ24193 through I2C. In particular, this writes a new value to the REG07 register. Looking at our firmware, I noticed that when the REG07 value is sent to the component, the REG07[3] bit is '0', while this is a reserved bit that must be '1' (which is obviously a mistake). Since the problem we encoutner is very rare I don't manage to reproduce it.

My question is the following: Is it possible that the wrong value of the reserved bit REG07[3] would explain that the BATFET doesn't alway open to switch off power ?

Additionnally, I noticed the following contradiction in the datasheet:

- Page 13 and 30, it is written that the watchdog timer is disabled with REG05[5:4]=00 (or at least that's what I understand).

- Page 25, it is written that the watchdog timer is disabled with REG05[5:4]=11 (or at least that's what I understand).

It would be nice to have confirmation of which setting disables the whatchdog, because if the whatchdog remains activated this could also explain my issue.

Thank you and best regards,

Thomas

  • 0
    TI__Expert 8230 points

    Hello Thomas,

    In addition to sending the REG07[5] command to turn off the BATFET, the VBUS voltage should be less than the UVLO threshold for the BATFET to turn off. The state of REG07[3] shouldn't affect the behavior of ship mode (even though it should be kept to 1).

    Another thing to pay attention to: ensure the I2C pull-ups are not collapsing before the ACK signal is sent by the device after receiving the ship mode command. We want to ensure the bq24193's state machine can respond to the command and not get stuck waiting for the ACK.

    Regarding the WDT, REG05[5:4] = 00 disables the WDT. We'll update the datasheet on the next revision. Ensure the WDT is disabled as REG07[5] will reset back to 0 if the WDT expires.

  • 0 in reply to Fernando Lopez
    Prodigy 50 points

    Hello Fernando,

                                  Thank you for those clarifications and quick answer.

    If I understand correctly, your first point might actually explain the behaviour that I observe. What happens if the BATFET turn off command is issued when the VBUS voltage is above UVLO, and afterwards (let's say 50ms after) the VBUS voltage drops below UVLO  ? Will the BATFET turn off as soon as the VBUS goes below UVLO, or the BATFET command is simply not latched, and thus the BATFET will never turn off ?

    Thanks,

    Thomas

  • 0 in reply to THOMAS PORCHEZ
    TI__Expert 8230 points

    Hello Thomas,

     

    THOMAS PORCHEZ said:
    What happens if the BATFET turn off command is issued when the VBUS voltage is above UVLO, and afterwards (let's say 50ms after) the VBUS voltage drops below UVLO  ?

    In this case, the device will wait for the VBUS voltage to drop below UVLO before turning off the BATFET (assuming the turn off command isn't cleared during the 50ms delay by the host or by the WDT expiring).

  • 0 in reply to Fernando Lopez
    Prodigy 50 points
    Hello Fernando,
    Thank you for your help, I managed to reproduce and understand my issue:

    Sometimes, when we switch to ship mode, our power supply is still connected. Thus, it is only when the supply is finally disconnected that the BATFET opens. The issue is that because of bouncing of the contact of the supply connector (DC jack connector), the disconnection is not instantaneous, and fluctuations of the VBUS happen. This can lead the BQ24193 to think that this is an adapter replugged condition, and it automatically turns back on the BATFET, while my system still thinks it is in ship mode. My system then stays stuck in a dead end waiting for its supply to go off, which never happens.

    Best regards,

    Thomas