• State
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 2 replies
  • Subscribers 64 subscribers
  • Views 102 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.

Original question:

BQ76972: Short circuit protection delay

BQ76972: Overcurrent in Discharge, DFET, and Watchdog Protections

Justin Reinhart
Prodigy 20 points
Part Number: BQ76972

Tool/software:

Hello,

I'm using the BQ76972 in a battery management system, and I'd like to accomplish two things, namely:

  1. Trigger a permanent fault and the fuse pin, due to a communication watchdog timeout at ~60 seconds, only if current is flowing through the shunt resistor.
  2. Trigger a permanent fault and the fuse pin, due to an overcurrent detected in the range of 100 to 500mS.

The fuse pin is my primary means of triggering protection, and I do not have FETs connected to the BQ chip. I am attempting to program the chip as follows:

  • Program permanent fail to activate the FUSE pin.
  • Configure 'Enabled Protections A' with the value 0xE4, as indicated in the manual on p. 40, to get 'fast' protection related to DFET. Only interested in OCD1.
  • Configure DSGFET Protections for OCD1 - DFETOFF_THRESHOLD_VALUE is -100, and DFETOFF_DELAY_VALUE is 0*
    • * - Not sure if DFET can 'instantly promote' or not (see questions) - have tried 0 and 1 here
    • Configure OCD1 threshold value very low (25 counts) so that I can trigger protections w/ desktop current supply
    • Configure OCD1 delay threshold to 125, or about 412.5 mS
  • Configure safety alert/status and PF alert/status settings to use OCD1

The goal of this programming is to get an OCD1 safety fault after ~412.5mS and disable the DCHG FET. I then want the DCHG FET to sense current flow "instantly" (or close) and cause a DCHG FET Permanent failure and trip the FUSE pin.

When testing this, I get both a safety alert and a safety status for OCD1 (good), but I cannot seem to promote these to a permanent failure and a fuse pin activation (bad). I've attached my complete list of register settings applied to the device, and voltage/current/temperature readings are functioning correctly as read through the I2C interface.

bq76972_lowsettings.h

Question for TI:

  1. Can DCHG FET protection be configured such that it 'instantly promotes' with current flow? If not, what is an alternate method to trip the FUSE pin on 100-500mS of current? I know the SCD operates in microseconds, and the permanent failures for current operate in seconds, but what's the middle ground?
  2. Can I instead use OCDL latching to instantly latch a fault? I've tried this with a counter delay value of 0 and 1, and with a latch limit value of 0 and 1, and I cannot seem to produce an OCDL fault.
  3. What is an appropriate method of promotion from safety status to permanent failure based solely on current?
  4. Is my 'watchdog-with-current' strategy in Goal 2 achievable? Should I be going about this another way?

Thanks,

Justin

  • 0
    TI__Mastermind 18035 points

    Hello Justin,

    Apologies for the delay and thank you for your patience. 

    Justin Reinhart said:
    When testing this, I get both a safety alert and a safety status for OCD1 (good), but I cannot seem to promote these to a permanent failure and a fuse pin activation (bad).

    I took a glance at this, and I may have missed it, but are you trying to use the Permanent Fail:SOCD protection? If so, do you mind sharing what parameters you have set for them? (Permanent Fail:SOCD:Threshold, and Permanent Fail:SOCD:Delay)

    Justin Reinhart said:
    Can DCHG FET protection be configured such that it 'instantly promotes' with current flow? If not, what is an alternate method to trip the FUSE pin on 100-500mS of current? I know the SCD operates in microseconds, and the permanent failures for current operate in seconds, but what's the middle ground?

    I will double-check if there is anything else that could be done, however, I think your current method is one way in attempt to do so.

    Justin Reinhart said:
    Can I instead use OCDL latching to instantly latch a fault? I've tried this with a counter delay value of 0 and 1, and with a latch limit value of 0 and 1, and I cannot seem to produce an OCDL fault.

    This should be possible. There is a chance the OCDL counter is recovering very quickly so it doesn't actually latch or it recovers from it. 

    Justin Reinhart said:
    s my 'watchdog-with-current' strategy in Goal 2 achievable? Should I be going about this another way?

    We have a host watchdog protection, however, we do not have a permanent fail (PF) protection associated with it. To clarify, do you mean to have a different PF trigger after a host watchdog timeout occurs?

    Best Regards,
    Alexis

  • 0 in reply to Alexis_H
    Prodigy 20 points

    Alexis,

    I attached my header file with all BQ76972 configurations to the original post, including the Permanent Fail OCD Threshold (1, intentionally low for testing) and Delay (125, or about 412 mS). Please reference this file for all conversation topics here.

    Re the DCHG FET: Thank you - please double check on this DCHG FET behavior, because this functionality may be critical to the functionality of the device incorporating the BQ chip. 

    The OCDL PF is not latching regardless of a 0 or 1 setting. I need to know the fastest conditions under which it can latch, and what setting to use.

    In my implementation, the communications watchdog is triggering a fault which should turn of the DCHG FET by configuration. The BQ should still measure current in this condition (as we are not actually using FETs) and trigger a DCHG FET permanent failure. In this way, the DCHG FET is our primary escalation method to get from 'fault' to 'permanent fail' in more than one case.

    Thanks for the feedback, and I'm looking forward to your response,

    Justin