• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 308 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.

TPS25750: Force using VBUS as power supply

Stefan Finauer
Prodigy 20 points
Part Number: TPS25750
Other Parts Discussed in Thread: TPS25751

Tool/software:

Dear TI E2E support team,

I am using the USB-controller's LDO_3V3 output pin to wake up my system when USB-C is connected. Once the connected MCU is awake, 3.3V power supply is provided to the USB-C controller's VIN_3V3 pin from outside and a patch is loaded through I2C. This works fine, but I'm having a problem shutting down when the USB-C device is still connected.

During system shutdown, the 3.3V supply is killed by the MCU which unfortunately causes LDO_3V3 to drop to zero as well. It seems like the TPS25750 then notices that VBUS is still present and powers LDO_3V3 from VBUS instead after a dead time of ~100ms. Problem is that this positive edge on LDO_3V3 wakes up my system again.

One possible solution would be to force the TPS25750 chip to use VBUS as supply for LDO_3V3 before killing the 3.3V supply from outside. This would mean that LDO_3V3 would remain constantly high and my system could shut down as desired (except TPS25750, which is OK). Is it possible to set the "dead battery flag" somehow through I2C or any other way to force LDO_3V3 to remain high in this scenario? Maybe even by loading another configuration patch? I measured that swapping power input between VBUS and VIN_3V3 doesn't cause any voltage drops in LDO_3V3 when no patch is loaded.

In case this is relevant: device Configuration using ADCIN1 and ADCIN2 is 'AlwaysEnableSink'.

Thank you very much in advance!

Fini

  • 0
    TI__Genius 12665 points

    Hi,

    We have received your question. TI US is closed for the Thanksgiving holiday until Monday 12/02. Please expect delayed responses. Our device experts will respond when TI US is back in office.

    Thank You

  • 0 in reply to Alex Liu
    TI__Mastermind 37855 points

    Hi Fini,

    As a general recommendation, we are asking customers to move away from the TPS25750 and to the TPS25751 where possible. It is a pin to pin feature compatible device. If you are not too far into the development process, it is strongly recommended. The issue would be the same with the TPS25751, switching parts would not resolve it.

    https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1346998/faq-tps25751-migrating-from-the-tps25750-to-the-tps25751?tisearch=e2e-sitesearch&keymatch=tps25751#

    Regarding the issue at hand, I'm not sure that the TPS25750 or 751 has a mechanism to control LDO3V3 in such a way. There is not a way to re-enable the dead battery flag using an I2C write. It would only be enabled if the device re-boots from VBUS which would require a power cycle.

    If I understand correctly, you need a signal that when unpowered is low, and will go high when the system is first powered from VBUS. If the system shuts down, but a device remains connected to the USB-C port, you are expecting the signal to remain high. The signal should only drop out when the port is disconnected.

    During system shutdown, is everything completely unpowered? I'm assuming there is another power input if you are clearing the dead battery flag?

    The issue is, if the Dead battery flag is not set and VIN3V3 is removed, the PD controller will power cycle and you will see VBUS, LDO3V3, etc. drop and the PD will need a new image loaded. The moment you reset VIN3V3 with the dead battery flag cleared, I do not think there is a way to prevent it from power cycling.

    Does the solution need to source voltages? Do you need to clear the dead battery flag?

    If there is another source of power that is stable during the shutdown, could you specify a shutdown sequence that ignores the reboot flag?

    I.E.

    SYSTEM SHUTDOWN STARTED

    (While in shutdown ignore LDO3V3 toggle, also may need to ignore PD interrupts during this time)

    Reset VIN3V3

    -> PD controller resets, and LDO3V3 is toggled, but is ignored

    -> PD controller now in a stable dead battery configuration, stuck in PTCH mode(no image loaded from MCU)

    Finish shutdown and undo the ignore.

    Thanks and Regards,

    Chris