Issue link: https://iconnect007.uberflip.com/i/1441485
80 DESIGN007 MAGAZINE I JANUARY 2022 Dunn: How did you f i r s t h e a r a b o u t semi-additive pro- cesses, and what made you want to learn more? Chase: In the semi- conductor indus- t r y, e v e r y t h i n g keep s shr ink ing and precision is paramount, so a few years ago we hit the lim- its of physics for what could be fabricated on a PCB. Back then, we tried a semi-additive process that had just been developed. While it was eventually successful, the turn times and communication with the vendor were poor at best. So, we kept searching for other solutions, and found that using a laminate interposer on a generic PCB got us by for a while. But alas, we eventually needed to remove the inter- poser as part of the signal integrity equation, so it was back to searching for a PCB-only solu- tion. To test, characterize, and evaluate our ICs, we must mount the bumped die directly on a PCB, so that means our ball pitch can be any- where from 100 to 150 mm apart. Yes, it's kind of humorous when you hear about designers dealing with 0.35–0.4mm pitch BGAs and our issues are tackling 100–150 mm pitch. We don't even deal in millimeters; we deal in microns. During the search, I came across Averatek and started communicating to learn more. We got in touch with a fabricator that was imple- menting the A-SAP process and did a trial run of the original board from many years ago. Low and behold, we came out with terrific results, had great communication with the fab vendor, and were extremely pleased with a quick turn- around on the fabrication cycle. We now have many boards using the process and continue to develop technology alongside the fab vendor. So, trace/space is just one part of the equa- tion, but since our ICs are typically RF, the topology of creating a squared-off trace (instead of the typically etched trapezoidal results of an etching process) is very beneficial as we con- tinue to increase frequencies. I got a bit carried away here, but there is a lot to be excited about. Dunn: Why do you feel that this technology is important to the industry? Chase: In the PCB industry, our current solu- tions can only take us so far due to the limita- tions of physics. I mean, PCB fabrication has remained basically unchanged over the past 50 years. Sure, there are lasers for smaller holes, and bigger and better presses for multi-lamina- tion cycles, but in the end, you still need to etch copper and plate in order to create your con- ductors. e whole etch process can only take you so far (or so small, if you will) that the limi- tations of the etch process become evident with very small or very dense boards. Another benefit is for RF performance. If you look at a cross-section, instead of having a trap- ezoidal trace (which will be more lossy), you'll see a squared-off trace with a much tighter tol- erance. is will be hugely beneficial as frequencies get higher and higher. D u n n : As you discuss t h i s t e c h n o l o g y w i t h other industr y exper ts/ customers that have a n e e d f o r f i n e f e a tu r e sizes, what stands out to you? Are they receptive? Figure 1: A comparison of shrinking features provided with traditional, mSAP and Averatek processes. Randy Chase