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.

TXB0104-Q1: What happens if input signal is high at power off?

Part Number: TXB0104-Q1

Hi team,

Our customer has an issue in TXB0104-Q1 just now.

It is used to translate 5V and 3.3V, Vccb is 5V and Vcca is 3.3V.

At power supply is off, when 5 V is applied to the pin B, Vccb seems to be  appeared the voltage slightly dropped from 5V.

Since it is off-state, I think that it will not break even if 5V is applied to port B.

Is it normal that the input voltage leaks to the VCC pin in the off state?

Since analysis has not advanced yet, I want to know if this situation is correct behavior.

Best regards,

Tomoaki Yoshida

  • Hi Yoshida-san,
    When the supply is off, what is the voltage on VCCA and VCCB pins?

    If VccA is ever larger than Vccb, problems can arise (there is a diode between the supplies, so always have Vccb > Vcca). I would recommend adding a pull-down resistor to the supply so that when it is in the 'off' state, the supply pins of the TXB device will be forced to 0V. In that event, the Ioff circuitry will prevent backflow into the device (an input or output can be driven to any valid voltage when the supplies are forced to ground).
  • Hi Emrys-san,

    Thank you for your reply.

    When the supply is off, both VCCA and VCCB are 0V.

    Because the power supply lines are cutted by switch, VCCA and VCCB are high impedance state.

    In this situation, 5V is applied to the B port, the voltage is confirmed at VCCB.

    I am think that this is not normal.

    Is this correct?

    Best regards,

    Tomoaki Yoshida

  • Yoshida-san,
    I am confused -- are VCCA and VCCB at 0V, or are they floating (high-impedance)?
  • Hi Emrys-san,

    I'm sorry, my previous comment is not correct.
    VCCA and VCCB is floating.

    Best regards,
    Tomoaki Yoshida
  • Hi Emrys-san,

    I think this situation is occured by body diode of high side PMOS, if it is as shown below.

    Thus, it is correct operation.

    Is my recognition correct?

    Best regards,

    Tomoaki Yoshida

  • The real problem here is that the supplies are left floating - if you can add a large pull-down resistor on those floating lines (ie 100 kohm) then they will be forced to ground and the device's Ioff protection will turn on, preventing all back-powering and current.
  • Hi Emrys-san,

    Thank you for your support.

    We understood the real problem you said and refer to your advice.

    First, as shown in the data sheet, when a pull-down resistor was added to the OE pin, the voltage at port B no longer passes through to Vccb.

    Is this effection of Ioff protection?

    The Vccb is still floating.

    I would like to know if Ioff protection worked just by pulling down the OE and whether it is a permanent measure for the real problem.

    Best regards,

    Tomoaki Yoshida

  • Hi Yoshida-san,
    The OE pin is not active when Ioff is working correctly.

    Since the OE pin is working, it seems that the device has power from some source. Forcing the OE pin low will put the outputs into high impedance mode as long as the device has power, and when it does not have power (ie Vcc < 0.5V) Ioff will keep the outputs in high impedance. This is a good solution if your application can support the pull-down on OE.
  • Hi Emrys-san,

    Thank you for your support.

    Since it is possible to pull down OE in this system, I will propose these measures to them.


    Best regards,
    Tomoaki Yoshida