• DocumentCode
    1945502
  • Title

    Module camber effect on card assembly and reliability for large flip chip BGA organic packages

  • Author

    De Sousa, Isabel ; McCormick, Heather ; Lu, Hua ; Martel, Robert ; Ouimet, Sylvain

  • Author_Institution
    IBM Canada, Bromont, QC
  • fYear
    2008
  • fDate
    27-30 May 2008
  • Firstpage
    397
  • Lastpage
    405
  • Abstract
    Component manufacturers and card assemblers are regularly confronted by the specification limits for the coplanarity of electronic components, and the effects of component warp age over the temperature range encountered during reflow processes used during card assembly. Card level assembly defects are often imputable to incompatible planarity levels of raw cards and their associated components, but the resultant impact of coplanarity mismatch is not well understood from the perspective of primary assembly yields at card joining and reliability. The present work studies the effect of module level camber on assembly yields and the reliability of a 55 mm x 55 mm, 2809 I/O FCPBGA component with lead free second level solder joints. The coplanarity of the components was modulated by applying different capping strategies, such as variation in heat spreader thickness, design, and attach process to achieve a range of coplanarties, including some well above current specifications, as measured at room temperature. Component coplanarity was measured for each component, and the warpage behaviour of each component variation throughout a reflow profile with a peak temperature of 240degC was also characterized. Module and card assembly limitations and the resultant solder joint geometries are further detailed in this work. Following assembly, the test vehicles were subjected to accelerated thermal cycling. The failures observed in the assemblies as a result of thermal cycling occurred in the second level ball grid array joints, located mostly in the center area of the package around the die shadow. The current study enabled the identification of the principal factors affecting the failure rates of the solder ball grid array, which include the board thickness, component design parameters such as heat spreader configuration and die size, and the shape of the solder balls, which is a function of the difference in height resulting from PCB and module warpage at the solidification temp- erature. Interestingly, room temperature coplanarity was not a strong indicator of reliability in this study. The work emphasizes the importance of achieving module to card coplanarity near solder reflow temperatures.
  • Keywords
    assembling; ball grid arrays; flip-chip devices; joining processes; reliability; solders; test equipment; accelerated thermal cycling; board thickness; card assembly; card level assembly defects; component coplanarity; component design parameters; component manufacturers; component warpage; coplanarity mismatch; die shadow; die size; heat spreader design; heat spreader thickness; large flip chip BGA organic packages; lead free second level solder joints; module camber effect; reflow processes; reliability; second level ball grid array joints; solder ball grid array; solder balls; solder joint geometries; strategies; test vehicles; Assembly; Electronic components; Electronic packaging thermal management; Electronics packaging; Environmentally friendly manufacturing techniques; Flip chip; Manufacturing processes; Soldering; Temperature distribution; Temperature measurement;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electronic Components and Technology Conference, 2008. ECTC 2008. 58th
  • Conference_Location
    Lake Buena Vista, FL
  • ISSN
    0569-5503
  • Print_ISBN
    978-1-4244-2230-2
  • Electronic_ISBN
    0569-5503
  • Type

    conf

  • DOI
    10.1109/ECTC.2008.4550002
  • Filename
    4550002