Abstract

Packaged LSI and hybrid devices used in high reliability military and space applications must pass a rigorous series of screens defined by Method 5004 of Mil Standard 883B. One of these screens is the Particle Impact Noise Detection (PIND) test. This test uses a very sensitive acoustic transducer to listen for particles within the package while the package is vibrated and shocked. We have used SEM, EDAX, and optical microscopy to analyze the particles from PIND failures. From these analyses we have identified the primary sources of PIND failures and have developed procedures that yield a low reject rate at PIND test. The device used in this investigation was a 1 K RAM die eutectically attached to a 24-pin leadless hermetic package (LHP). The package is solder sealed in a belt furnace with a gold-tin eutectic preform and a gold-plated cover. We have recovered the particles from PIND test failures by placing lead tape over a punched hole in the gold plated Kovar lid. The package is then vibrated until the particles pass through the hole and are attached to the adhesive on the tape. From the analyses we have identified many sources of particles that cause PIND test failures; the main source being the gold-tin solder preform used in the sealing process. We have investigated the effect of sealing materials, furnace temperature, furnace ambient, and package orientation on the number of gold-tin solder spheres.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call