• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 578 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.

UCD9090-Q1: State machine LGPO configuration understanding

Clément Letonnelier
Expert 2796 points
Part Number: UCD9090-Q1

Hello,

We are trying to configure one of the LGPO of the UCD9090-Q1 in state machine mode and so far we didn't achieve to reach our objectives.

Basically what we are trying to achieve is the following:

  • At power-up, our LGPO is expected to be at low level.
  • We expect the LGPO to change as when one of the GPI goes high.
  • Once it has changed due to the GPI, we expect it to change when one of the monitoring input goes in a power-good off state (which basically means our input power is gone).

We tried different configurations of both the AND PATH 1 and AND PATH 2 without luck.

Could you help us point how we should be configuring both AND PATH ?

Perhaps one of the misunderstanding comes from the initial state of the LGPO. Is the AND PATH 1 used AND PATH 2 when we go out of reset ?

Thanks,

Clément

  • 0
    TI__Guru 65700 points
    The initial state of the LGPO is dependent on state machine mode or not.
    If you have a state machine mode enable, only AND PATH1 is used to determine the inital state.
    For the other case, both AND PATH are used to determine the initial state.

    how about the case GPIs go low but rails are still good? shall LGPO stay or change with GPIs.
    Do you have any drawings of the signals?
    Regards
    Yihe
  • 0 in reply to yihe
    Expert 2796 points
    We are using state machine mode as my question was related to that topic.

    I could make a little drawing if that's unclear but basically our signal state change is the following:
    1/ 0 at power-up - monitoring input is not yet in power-good on and GPI is low
    2/ 1 once GPI goes high (it's a pulse of around 1 to 2 ms) - monitoring input is in power-good on at that stage
    3/ 0 once monitoring input is in a power-good off

    From a configuration standpoint, we could then go to 1 again if GPI goes high but our monitoring input going off means the power supply of the UCD is lost for good so next step will be a power-on of the UCD and obviously a reset of our state machine.

    Note that there could be some GPI going high after we reach state 2 but as we are expecting to wait for the monitoring input power-good off state, it is not important.

    To achieve that, I have put in:
    - AND PATH 1: the condition NOT GPI
    - AND PATH 2: the power-good of the monitoring input rail
    - Active high/Actively driven
    But whatever I am doing, the LGPO signal stays low.

    Clément
  • 0 in reply to Clément Letonnelier
    TI__Guru 65700 points
    if Your GPI's polarity is set to ACTIVE HIGH. you shall get a HIGH output after power up. The reason is that, GPI is low at power up(it is considered as DE_ASSERTED state since the polarity is HGIH). The AND PATH is NOT GPI. so when GPI is de-asserted state, NOT GPI make the logic TRUE. you shall get a HIGH OUTPUT unless your GPI's polarity is set to ACTIVE LOW.

    Please refer section 10.40.7 of www.ti.com/.../slvu352f.pdf about the state machine mode.
    Regards
    Yihe