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.

BQ76942: About OTP programming of PF at low voltages

Part Number: BQ76942

Hi!

We have a question regarding permanent fails during safety undervoltage.

We have a setup that looks like this:


SUV_THRESHOLD 0x92CB : 0x0708 (1800mv)
PROTECTION_CONFIGURATION 0x925F : 0x0732 ( PF_OTP and PF_FUSE is enabled)
( Say that we have SHUTDOWN_CELL_VOLTAGE 0x923F : 0x3E8 (1000mV) )

At the SUV level, the voltage is not high enough to be able to enable the fuse drive (with our fuse). Hence, we aimed to rely on OTP programing of the permanent fail to permanently disable the pack. Our problem is that as it seems, the permanent failed flag is not set if we recharge the cells back over the SUV limit. We have not performed any partial reset when starting up from shutdown this time...

In the documentation of the BQ76942 (11.3 Secondary Protections), we read "OTP programming may be delayed in low-voltage and high-temperature conditions until OTP programming can reliably be accomplished.". What does this mean?

With the given SUV threshold, can the OTP programing be performed? We have a feeling that even if so - it might take longer time than is available to us at these low voltage levels. 

Best regards

//Erik Almqvist

  • Hey Erik,

    OTP programming can only be accomplished when a voltage between 10-12V, unless your battery voltage is in that range at the time of an SUV, I don't think OTP will be possible. I will consult with a fellow engineer come Monday.

    Thanks,
    Caleb

  • Hi!

    Thanks for the support! This question is somewhat a follow-up question from this post:

    BQ76942: OTP write on permanent failure - Power management forum - Power management - TI E2E support forums

    In this post, conclusion were that there shouldn't be any requirements for writing PF to OTP. In the previous post, we tested PF writing by increasing the voltage limits to within our normal operation range, and that worked good. In these lower voltage regions, it seems PF is not successfully written.

    Best regards

    //Erik Almqvist

  • Hi Erik,

    What is your stack voltage in this case?  The device requires ~10V minimum for the programming to be successful.  For normal OTP programming of settings, we limit it to only programming within a narrow range, to avoid thermal issues with the associated circuitry if you were to attempt programming at a much higher stack voltage.  But for the PF case I recall we still allow programming at higher voltages, we just do it much slower to reduce any thermal impact.

    But if your stack voltage is below ~10V very much, the device probably cannot program the OTP cells, it would not have enough voltage to succeed even if it is attempting it.

    Thanks,

    Terry

  • Hi Terry!

    Thanks for your support! We have a 5p pack, so stack voltage is around 8V at this point. We know about the normal range for OTP, since we do program the OTP REGx during manufacturing.

    With your answer, I assume that there is no way for the AFE to write to OTP at this point (which is what we found). Do you have any suggestion on how to protect the scenario of passing the SUV, and then recharging the cells? Our best suggestion at the moment is to not rely on the OTP permanent failure indication of the AFE, but to store the permanent failure in the MCU. We kind of wanted to avoid moving logic from the AFE to the MCU, but we have problems getting OTP writes reliable?

    Best regards

    //Erik Almqvist

  • Hi Erik,

    So your stack gets to ~8V, you get the PF, but then if the device shuts down before the stack recharges to ~10V, I think the PF info would be lost.  When you powerup again, I assume the device will again measure the stack and get a new SUV, or is it starting to charge fast enough to avoid the SUV at that point?  

    If not, you may then need to store some variable in your MCU.  I'm not sure if there is another way around that.

    Thanks,

    Terry

  • Hi Terry!

    I agree this is a corner case, but at the same time I can't blame our tester for doing his job. As an example, this could be the case if someone attempted a direct charge on the cells under 8V. Normally, I agree the SUV would re-trigger again until depleted when woken again.

    We were simply surprised by the PF being cleared even after we attempted to program it, so we wanted to verify the behavior.

    Thanks for the explanation! It'll help us choose a route moving forward.

    Best regards

    //Erik Almqvist