Issue link: https://iconnect007.uberflip.com/i/811341
16 The PCB Design Magazine • April 2017 FABRICATORS SPEAK OUT ON HIGH-SPEED MATERIALS the parts look exactly the same." I'd say, "Aren't they supposed to?" They'd say, "Yeah," and then they'd giggle. It's process control. If a board shop doesn't have process control and they run panels through the etcher at different times for the same lot, if they run one core through the ox- ide multiple times but not the other cores in the same lot, if they laminate at different times for the same lot, if they run through the electroless and do some re-work, if we do all this variation to process, at the end there is not one panel that was processed like the other panel in that lot. That's the variation the RF guys just pull their hair out on. You've got to have process control from be- ginning to end. You want to have the finished product behave the same from board to board, panel to panel, lot to lot. Matties: Yeah, that's the ideal, is having them identical. But it's a rarity for sure. Partida: I don't want to forget; there are a lot of exciting laminate materials coming out that are processed like FR-4. Megtron 7, Megtron 6 is pretty good for a lot of RF. Megtron 7 will be great. Isola has I-Tera, Astra, Tachyon, that we're seeing customers using in the RF world. We don't have the processing issues of PTFE ma- terials where we have to use particular plasma cycles. It's low peel-strength and you can't etch it back. These new materials have almost the same performance as the PTFE but they're glass- woven and so they're more stable. We can control the registration. We don't have to go through difficult plasma cycles to deposit electroless copper in the holes. We can planarize the boards without distorting the panel. There are more selection and dielectric sizes in these new materials that are coming out, which are giving the RF community op- tions it did not have before because they had to use a certain bond ply in a certain location or a certain core in its location. Now they can use very different core thicknesses and use pre- preg and core interchangeably in a stack-up as needed. Matties: So, this offers a lot of flexibility? Partida: It really, literally opens up. There were things that required using a core and the regis- tration had to be perfect and they moved a lot, and there was just no way you could make it happen. Now you can just use a pre-preg and a foil and there's no chance of misregistering and you just register to the innerlayers through your precision registration software and you can dial in your drill into that one critical core and you can now do things you couldn't do before. This new material is pretty exciting. It's been out for about two or three years, which is new to the electronics industry but they seem very prom- ising. For some of the RF PCBs that have been built, I've heard excellent feedback from our customers, who say, "We built the first board and it worked, we matched the other boards to the same configuration and they all fired up exactly the same ..." Then when you respond with, "Doesn't this always happen?" And they say, "No, it doesn't always happen." Matties: You know it doesn't always happen when you ask that question; it's just fun to ask. Partida: And you want the feedback on the new materials too, right? Matties: Yes. We've talked to some of the material suppliers and one of their big issues is the length of time that it takes to get new material approved. They're saying it could be a multi-year process to get a new material approved into the workflow. How does that impact your material selections? We know that some of the material is being used before it's approved by Cisco or some of the recog- nized vendors. " These new materials have almost the same performance as the PTFE but they're glass-woven and so they're more stable. "