• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 908 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.

TM4C123BH6ZRB: GPIO Pulled Low During Startup

Akash Patel
Expert 6245 points
Part Number: TM4C123BH6ZRB
Other Parts Discussed in Thread: TM4C123GH6PGE

Hi team,

We are using the TM4C123BH6ZRB and during startup we are noticing that no matter if we configure a GPIO as input or output, the MCU is pulling the GPIO line low for about 200ms. Is there a way to prevent this from happening?

Regards,

Akash Patel 

  • 0
    Guru 117855 points
    You do not note if (anything) is tied to (that) GPIO - or if this, "pulling low" impacts many/most GPIO? Firm/I have past encountered this - but only when an interconnected, "Outside world" device, was able to power up far faster than the MCU.

    As always - has this effect has been noted upon a "reasonable" number of boards? Beyond this issue - does the remainder of MCU behavior meet expectations?
  • 0 in reply to cb1_mobile
    TI__Expert 6245 points

    I am wondering if the behavior of the device upon startup is such that the pins are pulled low no matter how the pin is configured. We are noticing this on all of the MCU pins.

    Regards,
    Akash Patel

  • 0 in reply to Akash Patel
    TI__Guru 72500 points

    Starting from reset, before the pins can be configured, they are in a high impedance state. They can float low or high based on leakage. It is possible that just the leakage through a scope probe will pull them low. If you must have the pins pulled high before the software can configure them then you need to add an external pull-up resistor.

  • 0 in reply to Akash Patel
    Guru 117855 points

    Akash Patel said:
    We are noticing this on all of the MCU pins.

    That answers (one) of my questions - offered in your aid.

    Silent is the "number of boards - or MCUs" so afflicted.    Has that exact MCU model been used in the past?    Are (other/past) such boards available for examination?

    Also missing - yet helpful - is any description of how you've measured this "pulling low" behavior.    Your experimenting w/different value "pull-up resistors" - I suspect - will reveal if the GPIOs are (really) driving to a "legitimate" low-level.   (i.e. use of a 1K6 "pull-up" will demand (near) 2mA to create an "honest" logic low.)

  • 0 in reply to Bob Crosby
    TI__Expert 6245 points
    Hi Bob,

    We just did some more testing on this on two of boards, both with the same results. We were able to test the PJ[0:7], PK[0:7], and PL[0:3] pins on our boards as those are the pins that already have external pull ups. PJ[0:7] and PL[0:3] behave as expected and stay high even through an MCU shutdown and power up, however, on PK[0:7], we are seeing these go low for about 200ms or so when powering up again. This is happening even though the only connection to the pin is the resistor connected to 3.3V. It seems like the MCU is pulling the PK[0:7] pins low for some reason but not the other pins that we were able to test.

    Any thoughts on what might be happening? Is there something special that we have to take into account for the PK pins?

    Regards,
    Akash Patel
  • 0 in reply to Akash Patel
    Guru 117855 points

    Checking our (private firm) notes - should this not be a confirmed (errata recognized) behavior - may we ask, "Where in your Port Initialization Order does Port_K appear? (i.e. first, middle, last) Could not hurt to alter that order - & test/observe.

    Should most (better still - all) of those Port_K pins appear upon "one side" (of the 4 sided MCU) our records reveal that a VDD pin (on that same side of the MCU) may have been "missed" (incorrectly routed) - or improperly soldered. (this effect is best observed during MCU Reset - and is likely to be "masked" once reset completes...)

    Let the record show that (earlier) you wrote that this, "unwanted GPIO "Drive to low"" occurred upon ALL GPIO pins.    We questioned that report - and note now that you report the effect to be "limited!"  (Port_K, only)

  • 0 in reply to cb1_mobile
    TI__Guru 72500 points

    I cannot repeat that behavior on the TM4C123GH6PGE part I have here. To clearly separate the hardware from firmware, would you please repeat the experiment with a pull-up resistor on PK0 and the part completely erased?

  • 0 in reply to Bob Crosby
    TI__Expert 6245 points
    Hi Bob,

    We had tried on two MCUs yesterday. One which had no code loaded onto it (fresh from the factory) and another with an empty project. Same results. Do you have one of the TM4C123BH6ZRB available to test on?

    We're seeing this happen even when we have the pins uninitialized.

    Regards,
    Akash Patel
  • 0 in reply to Akash Patel
    TI__Guru 72500 points
    Unfortunately I do not have a TM4C123BH6ZRB, however port K pins are exactly the same on the two devices. I would strongly suspect the board has those pins connected to something else, perhaps unintentionally.