Busy of the ADAS3024 is holden the low state at times

Hello.

Our product use a ADAS3023 and we check the ADAS3024 design verification.

When our product is powered on, the busy of the ADAS3024 is holden the low state at times.

We checked the ADI engineer zone for the issue and we found the improvement method.

For more detailed question information, see below(URL)

https://ez.analog.com/data_converters/precision_adcs/f/q-a/24654/adas3023-power-sequence/129074#129074

The PD/CNV/RESET sequence was uploaded on the question.(See below)

[Figure-1]

Therefore, we added the PD/CNV/RESET sequence[Figure-1 Additional Sequence] in our product firmware.

Consequently, the issue is improved and the busy of the ADAS3024  output correctly.

We want to know why the issue was improved by the additional sequence.

We think that the CNV sequence is very important for the busy issue.

Please tell us the detailed information on the connection between the additional sequence and improvement.

Best Regards

Takapoon



Adding Tags
[edited by: @skowalik at 1:17 PM (GMT 0) on 26 Feb 2020]
  • +2
    •  Analog Employees 
    on Feb 26, 2020 1:16 PM over 1 year ago

    Takapoon,

    The details as to why the sequence clears the fault is related to the internal conversion start logic.  Toggling PD high and pulsing CNV clears the state of an internal flop which inhibits additional CNV pulse registration until the current conversion is completed.  During some power up sequences this flop can become set in the inhibit mode before the internal timer and state logic in the core are ready resulting in a condition by which the flop is not cleared by the normal conversion execution.  This is the reasoning behind the strict power supply sequencing recommendations and why we often recommend powering up with PD and RESET tied to VIO through weak pull-ups.  if PD is tied high and CNV is accidentally pulsed there is no way for the internal flop to become stuck in the inhibit mode.

    Regards,

    Sean

  • Dear skowalik-san

    Thank you for your replay.

    We had understood the improvement method and close this discussion.

    Thanks

    Takapoon