Issue link: https://iconnect007.uberflip.com/i/1522641
34 PCB007 MAGAZINE I JUNE 2024 was a non-RF product. ere's been this con- vergence—sometimes accidental and some- times intentional—developing products that were originally designed for high-speed digi- tal applications, for specific RF applications by controlling the resin-to-glass ratio and control- ling the Dk. LaRont: I did purchasing and procurement for a period, focused on materials, and remem- ber when only "X" was available. There was no choice in the matter, and you factored it into your pricing. But now, with more choices, it's essential for the customer and the PCB fabricator to get the least expensive option up front, which includes buildability and yield. at's a key point. Most traditional CCL sup- pliers have good marketing departments, but their OEM market- ing folks are still most comfortable discuss- ing high-speed digital applications. RF design is a lit- tle different. You need people who can speak to the RF design- ers. at is probably the biggest challenge for conventional CCL sup- pliers entering the RF market. e RF guys have had years of experience working directly with RF product designers. RF designers have experience with RF products over many years. ey know how they perform. When they're looking at a new design, they are most com- fortable using what they've always used. Back in the day, you'd have the discussion with the supply chain folks saying, "You know, your product cost today is X dollars. is other product can save you Y dollars." But oen the savings did not justify the cost of evaluations and qualifications, engineering time, etc. But once that gap gets big enough, it is a differ- ent dynamic. Many of these new materials can provide more significant cost reductions. en everybody can rationalize the additional tests, etc., because there are so many cost savings to be had. LaRont: Between the traditional CCL supplier and RF suppliers, do the CCL guys derive any benefit based on the diversity of products they produce? To that point, another historical legacy is that many RF design layers were oen just a core. ose suppliers didn't have to invest a lot into developing a bond sheet or a prepreg to go with their high-speed material. at is something relatively new to the RF guys, whereas the con- ventional laminate suppliers have always had to make both laminate and prepreg. So, the tra- ditional CCL guys are a little bit ahead of the game in terms of developing prepregs or bond plies that go into multilayer RF designs. Shaughnessy: If you put your designer hat on as far as loss, Dk, and Df, what would be the cutoff where you think you could still utilize an FR-4-like material instead of a high-frequency material? I'm not a designer or an electrical engineer, so it's hard for me to answer that question other than to say, certainly the distance that you're operating at, and the frequency that you need, are some of the things that factor into it. If I'm just using my Fitbit, and I'm only going from my Fitbit to my phone, I don't need super high-per- forming material. I can get away with some- thing that's cheaper. When we are talking about other applications that have to be very high-speed and longer operating at distances, you will want to use something that has better control over Dk and lower loss. LaRont: With military and aerospace, every- thing is understandably reliability, reliabil- ity, reliability. Those organizations will be far more reticent to make a change because it takes so much time and money to change anything. I would assume this push is less likely to come from them. RF design is a little different. You need people who can speak to the RF designers.