• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 217 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.

BQ76942: Sensitive for commands during startup from shutdown

Adam Andersson
Prodigy 50 points
Part Number: BQ76942
Other Parts Discussed in Thread: BQ76952

Hi,
When waking up the AFE from shutdown, I poll the command 0x00 (Control Status()) in order to verify that the AFE is ready. However, I need to have a delay of around 350ms between waking the AFE up and starting to poll the command in order to make this work. If I don't have this delay, the AFE refuses to startup. I know that it takes approximately 200ms-300ms for the AFE to startup from shutdown. But should it really be that sensitive for commands at startup that it won't startup properly if commands are sent too early? Is a 350ms delay the way to go in order to make the startup sequence work properly or are there any other ways to do it?

Best regards,
Adam

  • 0
    TI__Expert 4085 points

    Dear Adam,

                  Generally, you can’t program to the chip until its powered up so you will have to wait that 200-300ms before it can be written. All the digital logic needs to be up and running before being programmed. There isn’t any exceptions to this. The 0x00 Control Status() command is used for legacy auto-detection, but it is generally not recommended for customer usage. Being in deepsleep instead of shutdown would help produce better results though if a fast response time is a necessity.

                  One possible way to make the startup sequence work better is to configure the ALERT pin to go high on the INITCOMP flag. This will drive the ALERT pin high as soon as the device initialization is completed and is meant to be used for an interrupt for an MCU. If this issue is only in SHUTDOWN mode, this would require the ALERT pin to be programmed in the OTP which is not always ideal. This would save a little bit of time, but you’d still have to wait until the chip is powered on.

     However, to get a better understanding I have a few questions:

    1. What communication protocol are you using to communicate?
    2. Will you be frequently going from SHUTDOWN to NORMAL?
    3. After sending the 0x00 message and failing to initialize, what state is the BQ76952 in? Does it go back to SHUTDOWN or something else?

    Best,

    Asher Nederveld

  • 0 in reply to Asher Nederveld
    Prodigy 50 points

    Thanks for your answer!

    1. I use 12C
    2. No that's not something that's done frequently.
    3. Yes, it seems like that's the case.

  • +1 in reply to Adam Andersson
    TI__Expert 4085 points

    Dear Adam,

    Based on that, I’d either include the 350ms delay before polling or use the interrupts on the ALERT pin to determine when to start since the device needs time to power. If it was frequently leaving shutdown, it would be worth using DEEPSLEEP instead, but since its not, DEEPSLEEP isn't necessary.

    Best,

    Asher Nederveld

  • 0 in reply to Asher Nederveld
    Prodigy 50 points

    Regarding question two, what is frequently and is there any known issues with shutdown?

  • 0 in reply to Adam Andersson
    TI__Expert 4085 points

    Dear Adam,

    There is not currently any issues with shutdown. Sometimes people use shutdown mode instead of sleep or deepsleep when the monitor is idling for short amounts of time. While there is nothing wrong with that, if you had done something like that it might be worthwhile  for you to avoid modifying the OTP memory and just using deepsleep if you wished to use ALERT interrupts. Shutdown is typically used for shipment or long term storage of devices. Hope that helps.

    Best,

    Asher Nederveld