• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 19 subscribers
  • Views 326 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

SN74AVC4T774: Termination undriven input required during power-up?

Ted Wolfe
Intellectual 260 points
Part Number: SN74AVC4T774
Other Parts Discussed in Thread: MSP-EXP430F5529LP, , SN74AXC4T774

Hello,

We plan to drive the inputs of the '774 from a Xilinx RFSoC after power up.  So there may be a short period of time (milliseconds) after power up when the '774 inputs are undriven while the RFSoC boots up and configures its output pins.  Is it necessary to provide pulldown resistors on the '774 inputs to make sure they are terminated and not floating even during this brief period of time?  Or is it unnecessary, nothing could possibly be damaged by floating the inputs for such a brief period?

thanks, Ted 

  • 0
    Guru 243180 points

    There are two problems that can be caused by floating inputs: cross currents, and oscillations. The first happens when the input voltage is not at the rails, and is relatively benign (especially if it happens only for a short time). The second happens when the input voltage is near the switching threshold, and can be dangerous. But for the input voltage to reach the switching threshold, the input needs to be allowed to float for long enough.

    This thread shows an example of the additional power consumption caused by seven floating input pins on the MSP-EXP430F5529LP board:

    It floats, but rather slowly. (There are no oscillations because the MSP has Schmitt-trigger inputs.)

    If the traces at the SN74AVC4T774's inputs are relatively short, and if there is not much electrical noise on your board or in your environment, then the voltage at the open inputs will not have an opportunity to drift by a noticeable amount.

  • 0 in reply to Clemens Ladisch
    TI__Mastermind 42600 points

    Hi Ted,

    On top of Clemens' inputs, you can also switch to the SN74AXC4T774 which should help in this scenario as well. It's been developed to improve power on performance eliminating any glitches.

  • 0 in reply to Dylan Hubbard
    Intellectual 260 points

    Thanks very much for the information Clemens and Dylan.

    Yes we can switch to the AXC if that's what you recommend to avoid power-on glitches.  I was planning to avoid such glitches by keeping the device disabled using the OEN pin until power is stable on both sides.  However, doing so may result in the inputs on both sides being undriven for a longer period of time.  I see that the AXC device datasheet contains the same warning that unused (or undriven) inputs must be held at GND or VCC to ensure proper operation.  I assume that this warning still applies, even if the device is held disabled by the OEN pin?  Therefore it seems to me that to be safe I need to terminate inputs on both the A- and B-sides of the device to VCC or GND at all times including when the device is disabled?  I think if the device were always enabled it would be adequate to only terminate the input side of the device, since the output side would be driven to a the same level, thus terminating the input buffer on the output side.  Is this reasoning correct?

    One more question:  We are also using the SN74AVC1T45DRL single bit transceiver.  Since this device has no OE input that can be used to disable the outputs until power on both sides is stable, I assume it is even more important to switch to the AXC version of the part with power sequencing glitch suppression?

    Thanks again, Ted

        

  • 0 in reply to Ted Wolfe
    Guru 243180 points

    If you want to be really sure, add a weak pull-up or pull-down.

  • 0 in reply to Clemens Ladisch
    Intellectual 260 points

    Thanks Clemens.  So I am correct that disabling the outputs using the OEN pin does nothing to mitigate the problems with floating inputs?

    -Ted

  • 0 in reply to Ted Wolfe
    Guru 243180 points

    The OE pin indeed affects only the outputs.

  • 0 in reply to Ted Wolfe
    TI__Mastermind 42600 points

    Hey Ted,

    Disabling the I/Os with the OE pin can cause issues for the input of the following device input since it will now be floating. If you don't plan on changing the directions of the channel, then just focusing on the inputs of the level shifter is okay (since as you said the outputs will always have a driven state when not disabled).

    If it is only the I/O you are concerned with floating, then the AXC774's glitch suppression circuitry should help keep those lines Low if there is nothing driving them so there wouldn't be any need for additional weak pull downs. However, the control inputs (e.g. Dir and OE) should always have a driven state. 

    To answer your other question, yes you are correct, the AXC1T45  will be the better choice in this situation since you will not have control over how the output reacts during power-up with the AVC1T45 (no ability to disable outputs)