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.

AM3359 Very High VDD_RTC current draw using TPS65910

Other Parts Discussed in Thread: TPS65910A, AM3359, TPS65910

Hi,

We are using an AM3359 and a TPS65910A.  We have a 32kHz clock on the TPS65910A and no clock on the AM3359 (we have RTC_XTALIN pulled low and RTX_XTALOUT disconnected per the datasheet).  We are, however, feeding VRTC from the TPS65910A to the VDDS_RTC on the AM3359 in order to make use of the PMIC_POWER_EN output to enable the TPS65910A.

We are observing a very high current draw (~350 uA) on the VDD_RTC path between the TPS65910 and the AM3359.  We really can't figure out where the leakage is coming from.  This is a serious problem as we would like to power the TPS65910A VBACKUP with a coin cell in order to maintain time and the current consumption is entirely too high.

Looking at the current measurements on the AM3359 wiki, http://processors.wiki.ti.com/index.php/AM335x_Power_Consumption_Summary, the VDDS_RTC current measured on the EVM  is fairly consistent (>300 uA) with what we are seeing.  This is not surprising as we based most of our TPS65910A/AM3359 interconnect based on the EVM.   Is this really what we should expect for an RTC power domain, or is there an issue with the EVM schematic?

We have tested this using TMX parts as well as Rev 1.0 Silicon.

-Mike

  • Hi Mike,
     
    A possible workaround is to supply VDDS_RTC from an external 1.8V LDO, drawing power from the main supply. One other thing is to stop the internal RC oscillator. Quoting from TRM section 20.3.3.2:
     
    "If the RTC is not used, the RTC_XTALIN pin should be held low and RTC_XTALOUT should be left unconnected. The RTC_disable bit in the control register (CTRL_REG) can be set to save power; however, the RTC_disable bit should not be cleared once it has been set. If the application requires the RTC module to stop and continue, the STOP_RTC bit in CTRL_REG should be used instead."
  • We recently found the RTC oscillator has an internal pull-up enabled when the oscillator is disabled, which is the default state.  This should explain about 150uA of the current you measured.

    We will are in the process of changing our recommendation for the RTC_XTALIN terminal.  The new recommendation when not using the RTC oscillator is to leave the RTC_XTALIN terminal open-circuit.

    Regards,
    Paul

     

  • Hi Paul,

    A VDDS_RTC current in Active Power is around 450uA.

    http://processors.wiki.ti.com/index.php/AM335x_Power_Consumption_Summary#Active_Power_2

    If the RTC is not used (OSC1 is disabled, both RTC_XTALIN and RTC_XTALOUT terminals are left the open-circuit, RTC_disable bit is set), is it around 300uA (minus 150uA)?

    The issue to increase the VDDS_RTC current is mentioned here: http://e2e.ti.com/support/arm/sitara_arm/f/791/p/333675/1290372#1290372

    Should the internal LDO not be disabled?

    Best regards,

    Daisuke

     

  • It is not clear what you are trying to do with RTC. Are you trying to disable the internal RTC?

    The RTC section of the AM335x Schematic Checklist describes the valid operating modes for RTC and defines external connections. Please review this information and make sure you are following these guidelines. Otherwise, you need to explain what you are trying to do and ask specific questions.

    Regards,
    Paul
  • Hi Paul,

    Thank you for your reply.

    Yes. The internal RTC is not used, so the internal RTC is disabled.

    When the internal RTC is disabled, what are the recommended connection and configuration? How lower is the power consumption than the case of enabling the internal RTC?

    For the recommended connection, should the connection of the "RTC feature disabled" in the RTC section of the AM335x Schematic Checklist be implemented?

    The issue to increase the VDDS_RTC current by disabling the internal LDO is mentioned here: http://e2e.ti.com/support/arm/sitara_arm/f/791/p/333675/1290372#1290372

    When the internal RTC is disabled and the OSC1 is disabled (both RTC_XTALIN and RTC_XTALOUT terminals are left the open-circuit), can the internal LDO be enabled?

    Best regards,

    Daisuke

     

  • Daisuke Maeda said:

    When the internal RTC is disabled, what are the recommended connection and configuration? How lower is the power consumption than the case of enabling the internal RTC?

    For the recommended connection, should the connection of the "RTC feature disabled" in the RTC section of the AM335x Schematic Checklist be implemented?

    Yes, please implement the "RTC feature disabled" column shown here:

    http://processors.wiki.ti.com/index.php/AM335x_Schematic_Checklist#RTC

    Please be sure to precisely follow the guidelines and ask a follow-up question if something is unclear.

    Daisuke Maeda said:
    The issue to increase the VDDS_RTC current by disabling the internal LDO is mentioned here:

    That's not terribly surprising.  Just because you're not using part of the device doesn't mean that no longer have to power it.  In that particular instance the lack of power to the RTC domain was causing a current path in the device that was drawing far more power than the LDO itself.

  • Hi Brad,

    Thank you for your reply.

    Best regards,

    Daisuke