• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 64 subscribers
  • Views 845 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.

BQ76930: Forcing reset on hung BMS without I2C

ST71353
Prodigy 20 points
Part Number: BQ76930
Other Parts Discussed in Thread: BQ76920,

With the BQ76930 (same applies to BQ76920), is there a way to manually force a reset on the IC without I2C?

For example, during some system transient states (let's assume these are unavoidable), the BQ IC has observed to get in a hung state, such that the input voltages are correct (after transient settle time) but the I2C is non-responsive. From the master (MCU) perspective, the lines appear to be reading high (i.e. not being held low), and manual bit bang STOP (SCL high, SDA low-to-high transition) does not recover. Unfortunately, I have not put an o-scope on this to confirm.

Cutting input power on BATT pin does not appear to work, since power is being fed back through a diode on VC10 and probably internal protection diodes on VCx pins (using EVM standard 1K resistors for filter). Due to alternate power paths, shunting the BATT pin to GND for a short duration doesn't seem like a good idea; or is this okay? Another idea is if a reserved function is within NC pin on IC that can be used to create a reset event. The only way to restore operation to BQ IC, is to physically remove the battery pack connection. This is not an option for field deployment.

Perhaps another way to look at addressing this issue, is if the BQ supply voltages fall below POR. According to datasheet, in very loud formatting, a "full device reset must be initiated by powering down both intermediate voltages (BAT-VC5X) and (VC5X-VSS) below Vshut and rebooting by applying the appropriate VBOOT signal to the TS1 pin". I'd like to create this condition without having the physically disconnect the entire battery pack, or adding a large FET switch network across all pack row voltages going in to the IC. If the later is what it takes, is there a recommended circuit I can evaluate?

Any guidance would be great, thank you.

  • 0
    TI__Genius 10830 points
    HI Steve,
    Thank you for your question.
    We are off for the week-end and will answer it when we're back at the office.
  • 0 in reply to David Hien
    TI__Guru 54715 points
    Hi Steve,
    Yes, you should put a scope on the lines to check their operation. There is no known hung state of the device. The normal method to clear a bus held by a slave is to send clocks until it releases the bus, the normal expectation is the slave holds the SDA line low so that the host can't control it and clocks are sent to shift out data from the slave until a high is sent.
    The bq76930 does not have a reset pin. Most of the part operates from the internal regulator stabilized by CAP1. Shorting CAP1 low momentarily removes power from the circuitry and causes it to re-start, possibly with a boot, but is not recommended as a reset.
    Removing all voltages (cells) and re-attaching is not practical in the field as you indicate. The design should avoid pulling voltages into the VSHUT region. POR is defined on the VBAT. We don't have a recommended circuit for disconnection.