Replacement of the drop test with the vibration test — The effect of test temperature on reliability

Due to the more extensive testing of new portable products the employment of efficient testing methods has become ever more important. The replacement of the commonly employed JESD22-B111 drop test with the vibration test has been investigated owing to the inconveniences related to the drop testing. The emphasis of the study has been placed on how the results of the vibration tests correlate with those of the drop tests at different temperatures. Comparison of the test methods was carried out by implementing the same experimental design with both testers and by using the observed failure modes and the number of load-cycles-to-failure as the evaluation criteria. The component boards used in this work were composed of the 12 mm x 12 mm Ball Grid Array (144 bumps, the pitch of 0.8 mm, and the bump size of 0.5 mm) and the printed wiring board (PWB), which was designed according to the JESD22-B111 standard. The experimental design consisted of three variables: i) composition of solder interconnections (Sn3.1Ag0.5Cu or Snl.lAg0.5Cu0.lNi), ii) PWB protective coating (Cu|OSP ja Ni(P)|Au), and iii) temperature of the component (23degC , 70degC, or 110degC). The failure modes observed in the component boards vibration-tested at the different temperatures were the same as those observed in the drop tests. However, copper trace failures were more common in the vibration-tested boards as compared to the drop-tested boards. In both the tests the performances of the component boards with different interconnection compositions and PWB coatings were very similar at the different temperatures. However, the relative decrease of the number of load-cycles- to-failure with increased temperature was larger in the vibration tests.