• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 3 answers
  • Subscribers 99 subscribers
  • Views 363 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.

66AK2H14: Reserved Pin shorted to Ground during assembly

Salman
Prodigy 60 points
Part Number: 66AK2H14


Hi Team,

Currently we are facing the issue in assembly of the 66AK2H14DAAWA24 part on the PCB.

As you can see in the X-Ray image AV37(RESERVED Pin) is shorted to AW38 (GND).

Can we operate the board in this condition without any malfunction since this in  reserved pin and suggested to leave unconnected will there be any problem if its connected to Ground.

In the design as well the board file the reserved pins are left unconnected but  during the assembly the lead has shorted with the adjacent ground pin as you can see in the encircled portion of the x-ray.

Please provide your suggestion.
what can be the appropriate method to do it, without the device getting damaged.

Regards,
Salman

  • 0
    Prodigy 60 points

    Attaching the x-ray image for your reference which is missed in the previous message.

    Here the short is observed only in the circled area, other dak patches are the decoupling capacitors present below the BGA.
    Please let me know your suggestions.

    Regards,
    Karthik

  • 0 in reply to Salman
    TI__Mastermind 46460 points

    Karthik,

    Pin AV37 is part of a reserved interface that is not being used.  Its pins are all open drain outputs.  As such, having this pin sorted to ground should not negatively impact this board for prototype functionality.  However, this is not acceptable for any board used for production as it violates the user documentation.

    Note that the presence of one short decreases confidence as their may be other opens or shorts that are not noticeable.  Do you see this on more than one board?

    You mention the other shadows from decoupling capacitors under the part.  I am surprised at the small number of capacitors under the device.  Since there are so many power balls under this device, every board that I have seen containing this device has had the entire backside under the device covered with decoupling capacitors.

    Tom

  • 0 in reply to Tom Johnson 16214
    Prodigy 60 points

    Hi Tom,
    Thanks for your reply, Yes this is the prototype and currently from the X-Ray we observed only one short in this pin and there is no other shorts observed.
    We have checked for the board impedance especially the core voltage, listed the same FYR.
    CVDD_0V95= 10E
    CVDD= 10.9E
    VDDALV_0V85 =35E
    DVDD_1V8 =606E
    DVDD_3V3 =1.5K
    Let me know if there is any problem in here or whether such low impedance is expected on the these CVDD and CVDD_0V95 lines?
    Yes actually there are many capacitor underneath the BGA but the ones which are here are bulk capacitors, attached the reference image underneath the BGA.

    Regards,
    Karthik

  • 0 in reply to Salman
    TI__Mastermind 46460 points

    Karthik,

    You are probably using a handheld ohmmeter to measure these resistances.  These meters us a 1.5V battery to apply a voltage to the board to measure the resistance.  They actually measure the current flow and calculate the resistance from the current measurement.  Since you are measuring across power supplies rated for less than 1.5V, the transistors are starting to power-up as the ohmmeter applies its test voltage.  This results in higher current draw thus resulting in lower resistance measurements.  The measurements reported are reasonable.

    Tom

  • 0 in reply to Tom Johnson 16214
    Prodigy 60 points

    Hi Tom,

    Thanks for your explanation.

    I will proceed with power on and further testing of the board.

    Best Regards,

    Karthik

  • 0 in reply to Salman
    Prodigy 60 points

    HI Tom, 

    Very sorry for the inconvenience,

    Actually i got above pin numbers from the assembly vendor which is wrong, actually today i personally went to verify the short or open and found that the actual short is between the pin number AV38 and pin number AU38 which is an VCNTL4 pin, now my question remains the same whether this short is acceptable?

    Currently we have given external pull up to 1V8 via 4.7K and also  connected to Current DAC. for smart refex controll for cvdd voltage.

    Please let me know how to proceed further, am extremly sorry for the wrong pin numbers.

    Regards,
    Karthik

  • 0 in reply to Salman
    TI__Mastermind 46460 points

    Karthik,

    Pin AU38, VCNTL4, is a pin required to operate the VCNTL interface which is part of the SmartReflex CVDD voltage control.  This device is not rated for operation without SmartReflex.  This is because the voltage cannot be set properly and the device may generate too much heat.  If this is a prototype that will only be operated in an office environment where you can provide additional airflow for cooling, then you should be able to operate the device with the CVDD supply fixed at 1.0V.  This option is not available for production boards.

    Tom

  • 0 in reply to Tom Johnson 16214
    TI__Mastermind 46460 points

    Karthik,

    How many prototypes have you assembled?  Do they all have this assembly error?

    Tom

  • 0 in reply to Tom Johnson 16214
    Prodigy 60 points

    Hi Tom,

    We have assembled 14 boards with 3 TI processor each in one board and I am facing this issue in one board only that too during the assembly of one processor.
    Yes we have heat sink as well as forced airflow card cage where these boards will be operated.

    Let me know for any further clarification.

    Regards,
    Karthik

  • 0 in reply to Salman
    TI__Mastermind 46460 points

    Karthik,

    Operation without smart-reflex may result in significantly more heat than the device is rated.  It should not be used in a production environment.  Also, it is on a board with multiple processors so mutual heating will also occur.  I recommend that you rework the board to get this short resolved.

    Tom