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.

TPS65982: IRQ behavior in SLEEP and IDLE mode

Part Number: TPS65982

Hi Team,

Could you tell me how IRQ behaves in SLEEP and IDLE mode? My customer would like to know whether the IRQ is also asserted  in those mode or not.

They plan to assign "High voltage warning" event to IRQ1 to check Vbus level easily on EC and they are expecting the function works well without any delay in those mode as well.

If there are any delays in these mode, could you tell me how much delay would be expected?

Regards,

Takashi Onawa

  • Hi Onawa-san

    Wake up from SLEEP to ACTIVE upon an I2C message is supported, however, the first I2C message is lost. To wake-up the device from sleep state, you would have to send one I2C command to the device to wake it up. This message however will not be used for anything other than waking the device. The device would have to already be awake for it to process the High Voltage Warning IRQ event.
    In IDLE state however, the device will still process IRQ events as the I2C interface is still active.

    Thank you,
    Eric
  • Hi Eric-san,

    Thanks for your prompt response on this.

    But I'm little bit confused because the datasheet says as below.

    5.4.3.2 Exit from SLEEP (Power Transition 02 – PT02)
    Exit from the SLEEP state is always to ACTIVE. Exit from SLEEP occurs because of the following events:
    • Any reset event
    • I 2C bus activity
    Any enabled interrupt event (I2C interrupt request, supervisor events, CC attach or detach events, and so forth)

    According to the sentence above, It seems that some of events make TPS65982 wake up.

    Could you tell me in detail what "I2C interrupt request" and "supervisor events" indicate?

    Regards,

    Takashi Onawa

  • Hi Onawa-san,

    When the device is in sleep state, it will lose the first I2C event and/or the first transaction on the CC lines. An interrupt event will always wake up the device and the device will process the interrupt. I apologize for the confusion with my initial response.
    However, if you are observing "High Voltage Warning" the device will always be in an active state when there is a PD contract negotiated. The only case when "High Voltage Warning" interrupt could be sent in sleep state is if a legacy device was putting 5V on VBUS with an implicit contract and it suddenly put more than 5V on the PD controllers VBUS.

    Thank you,
    Eric