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.

TPS2054B: USB Power Switch Odd Behavior.

Part Number: TPS2054B
Other Parts Discussed in Thread: TPS2065, TPS2054

We are having a very baffling issue with this USB Pwr switch. The design has been working for almost a year. On a recent PCBA build, the part doesn't supply power to "OUT3". Even if we lift the OUT3 pin into the air, it still hovers near ground. It also reports OverCurrent (OC3#) when there's clearly no current. To rule out effects of the hub disabling the switch ports, we disconnected those and installed pullup resistors to force the enables on. Although OUT1 is supplying power, the corresponding OC1 pin is asserted, which doesn't make sense (there are no USB devices attached during testing).

Strangely, our TPS2065 also shows weird behavior on a 5th USB port. With no USB load attached, the device reports over-current with nothing attached. We have replaced the device (U52 below) with a part from a different batch (new part from Digikey), looked at our Gerbers/ODB for any potential artwork changes (none found).

This behavior of reporting overcurrent, in itself, is not a show-stopper, and we may have always had this (we don't do anything with those signals), but the lack of USB power available on pin 11 is a serious problem.

Any thoughts?

  • Hi, Paul,

    As below functional block diagram shows,

    1. FET shutdown condition: (1) Vin UVLO (2) EN pin disable (3) Over temperature (4) Over current

    2./OC assert condition: (1) Over temperature (2) Over current

    As your description, no over over condition in your case, could you check

    (1) What's the temperature, is there over temperature condition?

    (2) What's the EN voltage? Does EN voltage touch the disable threshold? Is there big noise on EN pin?

    (3)What's the Vin voltage? Dose Vin touch the UVLO threshold? For Vin, I guess Vin should be okay, because you mentioned OUT4 can output voltage when OUT3 can't. Anyway, please check if Vin ripple is high? Big Vin ripple may influence the /OC logic.


    Regards,
    Bob

  • Bob,

    According to a thermal camera, and my finger, the switches are quite cool (room temperature), so there is no temperature issue.

    Vin is a clean 5V. The enables are also clean 5V, since they are (at the moment, at least) pulled up to Vin. There is no ripple or noise on any input.

  • Bob,

    According to a thermal camera, and my finger, the switches are quite cool (room temperature), so there is no temperature issue.

    Vin is a clean 5V. The EN inputs are also clean 5V, since they are (at the moment, at least) pulled up to Vin and disconnected from the hub. There is no ripple or noise on any input.

    I should add that the OC outputs (on both TPS2054 and TPS2065) have a timing component to them that is interesting. From power up, OC1 is immediately low. From power-up, OC3 starts out OK, then goes low after 2 seconds. On the TPS2065, OC starts out OK, then goes low after 10-12 seconds. This is very repeatable. For example, even after a very brief power cycle of TPS2065 the OC pin doesn't assert (low) until 10-12 seconds.
  • Hi,

    This sounds pretty weird. Have you find the cause?

    Do you verify whether the part damaged? And where did you get the part?

    Regards,
    Bob
  • This is a bit embarrassing, but we hadn't noticed a few signal outputs unconventionally leaving the left side of the schematic sheet. Those Overcurrent signals fed a processor board who's "inputs" got misconfigured as outputs, and thus they stomped on the OC signals. The USB power switches work fine as long as other circuits don't interfere with them. Thanks for your support.