Here is the abstract you requested from the MIL_2010 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Low Cost Hermetic Sealing of Large Electronic Packages|
|Keywords: Hermetic Sealing, Projection welding, Electronic packaging|
|As one investigates the benefit/cost ratio of various packaging options, they will be impressed by projection welding of platform style packages. In fact, for products that can use axial leaded packages, it continues to be most robust and cost effective hermetic sealing technology ever developed. Our vision has always been that the same process used for sealing smaller TO-packages was expandable into a reliable, cost effective method for sealing much larger packages. The first obstacle we encountered was the availability of equipment capable of supplying sufficient weld current to weld larger packages. In 2003, we were granted a patent on a new welder design. Our first series of welders are compact machines that deliver about 150,000 amperes of weld current, enough to seal 8-10 linear inches of package in a single discharge lasting a few milliseconds. The technology used to build this welder is capable of delivering higher outputs of 500,000 or even 1,000,000 amperes. This new welder technology was immediately accepted by industry, but as larger packages were submitted for welding, an unexpected limitation began to appear. When welding 6.0 linear inch packages, we began to notice leaking welds, usually near the corners of square or rectangular packages. This was an unexpected complication, and we set about trying to identify the root cause of these leaks. One of the tools employed in our investigation was an imaging focal plane array (FPA) to measure and record the peak temperature encountered during the welding process. We noticed that leaking welds always exhibited lower peak temperature near a corner. Upon measuring the projection height around the package perimeter, the reason for this observation became obvious. Weld projections are manufactured by the coining process. Coining can result in non-uniformity of the projection height. The resulting height non-uniformity tracked precisely with the temperature fluctuations we observed during welding. In this paper, we will show some of our test data, and illustrate steps we have taken to mitigate the problem. Although the jury is still out on projection welding of very large packages, the data collected to date suggests that 20.0 linear inches should be easily achievable.|
|Thomas E. Salzer, President