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.

LMC8101: DSBGA solder joint failures

Part Number: LMC8101

Hello, we are attaching this device to an FR4 substrate and experiencing a small % of failures that appear to be caused by a poor connection to the input pins B3 and C3.  We recently cross sectioned a failure to find out what is happening.  The picture below shows cross section photo of pin C3 when it is soldered to the board.  There appears to be a fracture in the solder joint near the pad of the LMC8101.  

My question is regarding the fracturing of the solder joint, is this something that has been seen before and is it known what the causes could be?  Some more details about our pad geometry: we have10mil pad sizes and 1 to 2 mils larger mask openings.

  • Ken,

    The data sheet example footprint pattern uses round 0.16mm = 6.3 mil pad and 0.05 mm = 2 mil (max) solder mask spacing

    The larger pad could be part of the issue also the mask opening looks bigger than 1-2 mil

    The solder is still on the pad, therefore adhesion is good. I don't know the source of the stress that caused the fracture.

    I will request that our solder expert will also take a look at this.

  • Ron, thanks for your response and for passing along my question. You are right, the mask opening does look quite a bit larger than I thought.
    -Ken
  • Hi Ken,

    See AN-1112.

     http://www.ti.com/lit/pdf/SNVA009

    This packaging appnote was released around the time the LMC8101 was released (and referred to in the original National datasheet) to address difficulties customers were having at the time of this "new" package.

    Also see the other appnotes in the DSBGA section of the packaging page:

    http://www.ti.com/support-packaging/packaging-resources/SMT-and-application-notes.html#dsbga

    It is critical to get the pad size opening correct. If the pad is too wide, the solder ball will wick off the die and onto the PCB pad and spread-out, creating the void at the chip. They recommend that the pad surface be 2/3 the side of the ball diameter.

  • Hi Paul,

    Thank-you for your response.

    Looking at the information you sent, I agree that our designed pad size is larger than recommended.  We are ~ 254um and the recommended is 145um to 185um per the Do's and Don'ts table in AN-1112.

    There was also the graph on page 6, "Figure 4. Impact of PCB Pad Size on Reliability for 0.17mm Bump Package", showing decreasing reliability with larger pad size.

    Regarding the cross-section photo I showed, do you think the "fracture" is really a void from the original solder attach process or could that fracture have been caused by some later physical shock? Do you have any idea if that fracture is consistent with the type of failures found in TI's reliability testing?

    I'm just looking for some confirmation of what we are seeing.

    Thanks for your help!

    Ken