• State TI Thinks Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 2646 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.

TXS0108E: output keep low when disable (OE=low)

CK Ho
Expert 1780 points
Part Number: TXS0108E
Other Parts Discussed in Thread: BQ24715

Dear sir,

Our customer is using TXS0108E, and occurring a problem,

When OE pin is low, the output should be Hi-Z, right?

But they found there a few IC that pin14 will pull the voltage level low when OE is low. It will cause their system cannot be turned on.

And the failure follows the suspect IC. Replacing the IC can resolve the issue.

Customers want to know the detail circuit in TXS0108E. Why is the phenomenon happen?

Would you please help to provide the internal circuit of TXS0108E? Thank you very much.

  • 0
    Guru 316690 points

    Please show oscilloscope traces of the ACCEL_INT and ACOK signals when a working chip is switching.

  • 0 in reply to Clemens Ladisch
    Expert 1780 points

    Below is the waveform fo the ACOK and ACCEL_INT from pass unit.

    system off -> TXS0108E disable -> ACOK=ACCEL_INT=low -> then plug in AC -> ACOK become high -> system turn on -> TXS0108E enable -> ACCEL_INT become high.

    This is the correct process. But if using failed unit, ACOK and ACCEL_INT always keep low.

  • 0 in reply to CK Ho
    Guru 316690 points

    In the schematic, the ACOK off-sheet connector points outwards, but the description implies that some external device drives ACOK low. Does this imply that the schematic is wrong?

    At the beginning, both ACOK and ACCEL_INT are low. How can this be possible if OE is low? Are you sure that OE is low?

    What exactly is connected to ACCEL_INT and ACOK? When exactly do other devices drive these signals high or low?

  • 0 in reply to Clemens Ladisch
    Expert 1780 points

    Dear Clemens Ladisch,

    1. Yes, I also think the schematic is wrong. The arrow on the ACOK should be left-pointing.

    2. Yes, the OE is low. There is a pull down resistor on OE pin. The OE will be driven to high when system turn on.

    3. The AOK signal is connected to a battery charger IC, BQ24715, pin5. When AC adapter plug in, this signal will become high.

  • 0 in reply to CK Ho
    TI__Mastermind 45725 points

    Hi CK,

    Can I see the same waveform but with the OE signal probed as well. And then can you also replicate the same scope capture with a device that isn't working? I'll see if I can see anything going wrong, but so far nothing stands out.

    I would also look into starting an FA (failure analysis) with your distributor or dedicated sales support. This will allow us to get the problem devices here and test them which should give us more insight to the root cause of the problem. 

  • 0 in reply to Dylan Hubbard
    Expert 1780 points

    OK, I will discuss with our customer.

    Thank you very very much for your support.

  • 0 in reply to CK Ho
    TI__Mastermind 45725 points

    Hi CK,

    One more thing I want to add since I didn't catch it before: channel 1 and 2 (A1/B1,A2/B2) don't need pull-ups. Strong pull-ups tend to give these devices issues since they already have internal pull-ups. I don't think it is related to the issue you are having, just wanted to bring it to your attention so you can save space on the board.

  • 0 in reply to Dylan Hubbard
    Expert 1780 points

    Hi Dylan Hubbard,

    Strong pull-up means large pull-up resistor or small pull-up resistor??

    The customer said they have a 33K pull-up resistor on ACOK. If they change the pull-up resistor from 33K to 10K, the issue can be solved.

    They want to know why the pull-up resistor will impact the TXS0108E behavior, and why it cause this issue??

    Thank you.

  • 0 in reply to CK Ho
    Guru 316690 points

    "Strong" means more current, i.e., a smaller resistance.

  • 0 in reply to Clemens Ladisch
    Expert 1780 points

    Yes, I also think strong pull-up means smaller pull-up resistance.

    But now the issue can be solved by reducing the pull-up resistor.

    Would you please help to explain why the pull-up resistor will impact the TXS0108E behavior, and why it cause this issue??

    Or can we submit a FA request??

  • 0 in reply to CK Ho
    TI__Mastermind 45725 points

    Hi CK,

    What happens when the pull-up is removed? The battery charger pin 5 requires a 10 k pull-up since it's open drain, our device has an internal 10k pull-up so an external pull-up won't be needed. However, based on your previous description of the operation sequence the TXS device will still be disabled when ACOK is supposed to go high. This means the internal pull-ups will be disconnected and the line won't be pulled high until our device is enabled, but i'm not sure if that will be an issue or not.

    As far as the issue, there is something pulling that line low. Since you said the issue follows a certain device from board to board, it leads me to believe there may be damage to the part causing that pin to short to GND.  However, if a stronger pull-up is actually fixing the issue with the problem device then this may not be the case anymore. I can't really say for sure, but i'm skeptical since you still have other boards working with the same circuit.

    I would still pursue an FA, however I'd make sure there is nothing else that can pull the line low. This means the battery charger should have it's conditions met for it not to drive the line low. Also I would check whats connected to Accel_int. There may be pull-downs externally on that line or even internal to the device using the interrupt signal.

  • 0 in reply to Dylan Hubbard
    Expert 1780 points

    Dear Dylan Hubbard,

    Now we found the root cause maybe is leakage of TXS0108E.

    For the suspect IC, there is about 225uA leakage current through into Pin14 at 500mV. (OE is low and Vcc=0V)

    But there is about 180uA for another pass IC.

    We have submitted a FA request, the number is CPR191032731.

    Thank you very much.