• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 99 subscribers
  • Views 735 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.

nRESETIN_OUT not de-asserting

Thomas Wesenberg
Prodigy 110 points

We have the Sitara on a piggyback module.

When this module is on one type of mainboard, the Sitara starts up as normal.
Is the module on another type of mainboard, signal nRESETIN_OUT is not de-asserted, although PORz ist deasserted and CLKOUT1 is starting too.

We are wondering what could be the reason for the Sitara not deasserting nRESETIN_OUT.
This signal is not touched by other components, it just has the pullup resistor. Voltage for this resistor goes up.

  • 0
    TI__Guru**** 161600 points
    Hi,

    Could you share the relevant part of the schematics?

    Best Regards,
    Yordan
  • 0 in reply to Yordan Kovachev
    TI__Guru 62855 points
    I assume you are using the same module that previously worked on the other mainboard. If so, the problem appears to be on the mainboard.

    Check all power supplies, clock sources, and reset sources. The non-working mainboard must be missing something very basic if the device does not power-up and release nRESETIN_OUT.

    Regards,
    Paul
  • 0 in reply to peaves
    Prodigy 110 points

    The problem is 100% reproducable with one module, the other 49 modules work without problems.

    We want to start mass production but need to have an explanation for this effect and to find a solution before doing so.

    Could you state which signals are relevant for releasing nRESETIN_OUT?
    5V power is going up. Other supply voltages are generated on the module and Pwronrst is released, Sitara working clock on OSC0 is starting. But no release of nRESETIN_OUT.

    The schematic is available here, it's quite similar to the beaglebone: http://www.guck-mal-hier.de/e2e.ti/AM3352.pdf

    Here are two oscilloscope measurements of the module located on different type of mainboards, first is ok, second is faulty.

    http://www.guck-mal-hier.de/e2e.ti/PM-ok.png

    http://www.guck-mal-hier.de/e2e.ti/PM-nok.png

  • 0 in reply to Thomas Wesenberg
    TI__Guru 62855 points
    This issue does not sound like a design problem if 49 out of 50 boards work. You will need to contrast the differences in this non-working board to the 49 working boards to understand what is different on this board. It sounds like you may have a PCB assembly problem on the non-working module. This could be hard to prove without removing the current device, re-balling, and re-installing. Even this may not prove anything if it still doesn't work. The best way to prove you do or don't have an assembly problem is a destructive test called Dye and Pry. You can google "Dye and Pry" for more information.

    Have you removed all flux and thoroughly cleaned the board after assembly to insure there are no high leakage conductive paths between pins?

    Applying power to all the power rails as defined in the data sheet, applying a valid reference clock, and releasing power on reset is about the only thing that needs to be done for it to release nRESETIN-OUT.

    Make sure you have the EMU[1:0] inputs pulled high.

    Regards,
    Paul