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.

AM3352 wake up delay from RTC_ONLY sleep condition

Other Parts Discussed in Thread: TPS65217, AM3352

Hi all,

We are working with Beagle-bone black & testing different sleep modes.

When we put the AM335x into RTC_ONLY sleep mode, all the LDOs & DC-DC regulators of PMIC except VRTC (LDO1) will be turned OFF.

According to Figure 8-5. of TRM, only way to wake up the processor during RTC-only sleep mode is through the WAKEUPn signal. WAKEUPn signal can be triggered by pressing the PB_IN switch connected to PMIC.

During testing, we observed that there is around 6 sec delay between pressing of PB_IN switch & board rebooting. We probed the PB_IN , WAKEUPn & PWR_EN signals & below are the observations.

1) There is 58mS delay between PB_IN switch press & WAKEUPn going LOW.

2) There is around 5.38S delay between WAKEUPn going low & PMIC_EN going high.

Please find the below snapshots for details.

We are unable to find out the reason behind delay between WAKEUPn & PMIC_EN signals. Is this behavior expected ? Please confirm.

Thanks,

Suresha N S

  • Suresha,

    This behavior is expected; the nWAKEUP pin is delayed 50ms over the power-up event and will remain low for 50 ms after the PWR_EN pin has been asserted. The PWR_EN pin has a 5 second time window in which it waits to be pulled high. Please see the nWAKEUP section of the TPS65217 data sheet (SLVSB64G).

    Janice  

  • Janice,

    Thanks for your response. 

    The datasheet says  delay between WAKEUPn going LOW & PMIC_PWR_EN going high to be 5 sec max. We are observing around 5.8S. 

    For our application, we cannot have a wakeup delay of 6sec. Is there a way to reduce this delay ? Please suggest.

    Regards,

    Suresha N S

  • Janice,

    Our requirement was to poweroff the board using SW command.

    Earlier we had tried to achieve the same by putting AM3352 into RTC only sleep mode. In this case, the RTC LDO VRTC will be ON.As detailed earlier, we were observing around 5.8sec delay during wakeup from  this mode.

    We thought of powering OFF PMIC completely as an alternate option since we are not planning to use the AM3352 RTC. Thus we set the OFF bit of status register of PMIC before entering Poweroff. 

    With this configuration, when we execute Poweroff command, PMIC will get turned OFF completely i.e. even the VRTC voltage will be 0V. During this case, the AM3352 boots up on a single click of PB_IN switch without any delay.

    Since this meets our requirements, we are planning to go ahead with this approach. 

    Regards,

    Suresha N S

  • Sounds like the RTC is failing to respond to the wakeup signal. Check the EXT_WAKEUP_* settings in the RTC_PMIC register.

    Also, what AM335x revision are you using? Revision 1.0 has a known erratum that is relevant here (see advisory 1.0.5) which makes RTC-only mode useless.