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.

TPS65217: Brownout issue

Part Number: TPS65217

The application report SLVA901 address a fault due to Vin brownout when there is no battery.

Can you go into more detail of what this fault is?

I assume the fault is that PMIC SYS output turns off when a brownout occurs but does not come back up upon Vin recovery.

My circuit is like Case A in SLVA901 and when Vin drops below 3.6 Volts the PMIC turns off and does not turn back on when VIN is returned to 5.0V.

I tried solution circuit number 2 and the same thing still happens. I had assumed that the PMIC would turn back on when the Vin voltage recovered.

Is this because the brownout lasted more than 1 second?

  • Greg,

    I've assigned this post to the product owner, he can help answer your questions. I apologize in advance that there may be a delay due to Thanksgiving holiday in the US.

    Regards,
    Karl
  • Greg,

    I would need to see a schematic and a scope shot of the issue occurring in your application in order to debug the problem.

    My guess would be it appears on the surface that you have implemented the Case A circuit but you are actually working with a Case B circuit, but it is impossible for me to say with confidence unless I am looking at a schematic and scope shots.

  • I do not have any scope shots, just a report that none of the PMIC outputs come back up
  • It appears that this started as the BeagleBone Black schematic that has been marked-up. I cannot tell where VIO is actually connected in your design, but I will assume it comes from LDO1 and is not generated from a different source. When there is no battery & VIO is generated by the TPS65217 (and not applied externally), then this will be Case A in the App Report.


    When you used a resistor to connect BAT to GND with a light load, was BAT_SENSE shorted to BAT1/2 (TP6 to TP5)?

    Having BAT and BAT_SENSE shorted together is important for the TPS65217 to recognize that no battery is actually connected. This is clearly shown in Figure 7 for Solution Circuit 2 (page 9).

    This is the only suggestion I can make at the moment without seeing any scope shots showing the issue occurring.

  • VIO is generated by VLD01 (look for R7 near pin 3).

    BAT_SENSE is shorted to BAT1/2 (TP6 to TP5) and the two pins are taken to DGND through a 1k ohm resistor.

    What scope shots would you need?

  • Greg,

    A reproduction of Figure 8 on page 9 of the app report would be helpful, with AC, SYS, & BAT pins being monitored. It may also be helpful to scope BAT_SENSE. If BAT_SENSE is identical to BAT, then PB_IN could be scoped.

    These images were drawn based on scope shots I captured while testing the TPS65217EVM to test the issue and solutions.
  • How does this resistor value interact with the battery detection current of 10mA as described in the TPS65217 data sheet, sections 9.3.9.1 and 9.3.10? Would it be better to use a lower value such as 100-270 ohm to keep the BAT voltage under 3V or even trigger the short circuit behaviour?
  • When TS pin is left floating, Battery detection does not occur.

    What you are seeing during power-down on BAT pin is a small amount of current from the SYS pin charging the BAT pin when they are connected through the power path. This produces a voltage that is quickly discharged by a 1-10kOhm resistor.

    However, you can use a stronger resistor if it is required by your application. Under normal operating conditions, the BAT pin will not be connected in the system when TS is left floating and the resistor will not add additional load to the main power supply.