Issue link: https://iconnect007.uberflip.com/i/1484263
12 PCB007 MAGAZINE I NOVEMBER 2022 Todd Brassard: You might call it innovation for the United States, but the truth is that UHDI is a 20-year-old technolog y, and the U.S. is playing catch-up to offshore man- ufacturers. When I first tried to understand what a substrate was, much of my initial insight and perspec- tive came from a 1990 doc- toral thesis titled some- thing like, "Quality con- cerns in manufacturing flip chip assemblies with organic substrates." is was a well-formed docu- ment from over 30 years a g o . I t h o u g h t , " H o w could the U.S. PCB industry have its head in the sand for 30 years?" I suppose we all know the answer to that question at this point. Happy Holden: During that period, I heard Nan Ya spent $400 million on a new plant to make these substrates. Whoever heard of a PCB shop that cost $400 million? All class 100, 11 stories tall, with AOI machines costing $1.5 million each. Brassard: Some amount of capital investment is unavoidable to get into substrates, but for certain types of substrates, PCB manufactur- ers have 90% of the capital equipment needed, although one discovers quickly that UHDI capability is a prerequisite for most, if not all, substrates. Johnson: Can you convert a PCB shop into a substrate shop, or do you need to build it to be a substrate shop in the first place? Brassard: Before you can answer that question, you must understand that there are a wide range of advanced package constructions and associated types of sub- strate with many levels of complexity; some are definitely doable by U.S. PCB manufacturers with minimal capital invest- ment, but more com- plex designs need signifi- cant CapEX investment. What level of technology does the U.S. require with respect to building sub- strates today, tomorrow, in the coming decades? e right PCB manu- facturers with the right e q u i p m e n t s e t , e n g i - neers, and workforce will be able to stand up solu- tions to meet some per- centage of the industry's immediate needs in a relatively short term. But I am confident that a subset of U.S. PCB manufacturers, in time, can and will catch and surpass the capabil- ities of offshore manufacturers—not the capac- ities or the large-scale economics, but the low- volume incredibly advanced technologies. e road is already forming before us, the U.S. will close the gap, push past the narrow definitions of SOTP and SOTA, and get back to innovating. Let's keep in mind that as microelectronics continue to shrink, more of the design of a sys- tem will be built on a substrate as opposed to a PCB. Meredith LaBeau: As technology pushes to min- iaturization and heterogenous integration it seems natural that the printed circuit boards of today will be the substrates of the future. e need to fan out technology will drive the build-up concept as seen in substrates. We can already witness the technology drive from HDI PCBs to UHDI build-up PCBs and substrates. is could potentially be manufacturing tech- nology that resets the technology curve allow- ing for new manufacturing innovations. Todd Brassard