Issue link: https://iconnect007.uberflip.com/i/1520701
14 DESIGN007 MAGAZINE I MAY 2024 But lo and behold, you can target the same Dk values that people were getting in tradi- tional Rogers materials with the very-low and ultra low- loss materials. Isola has a product that was developed for applications similar to Panasonic Meg- tron 6, and they have given me permission to talk about it. It is called I-Tera MT40. It performs in that very low loss high-speed digital range, w h i c h i t w a s d e v e l o p e d for, but, with a copper foil that's a little bit smoother than what can be used in some RF materials, you can get to the required RF performance levels as well. e middle col- umn of this image shows high-reliability, high-speed digital (HSD). You can see this I-Tera MT40 product is also called out for RF microwave products. e point is that as the historical FR-4 folks develop better prod- ucts, they can certainly be used in the RF world. LaRont: What is the cost or cost differential? Are we reaching a point where the FR-4 materials with these properties will approach the cost of RF materials? Kelley: ere's a range, of course, from different suppliers and different product configurations. ere is a lot more low-Dk and low-Df glass on the market today, which is a factor in pricing. Certainly, the lower your Dk and Df levels, the more your price will typically increase. Moyer: I think there's a second price-related or costing-related issue, and it's not just that there might be slightly higher cost on the material itself. But it's that non-PTFE material is simply easier to work with in manufacturing, and you can cut out certain processes, likely improving yields. at is the total cost of the product vs. just the raw material. Kelley: Also at Isola, there is a product called Astra MT77, which is not a PTFE- or flu- oropolymer-based product, but it performed like that in many applications. It could compete with, for exam- ple, the Rogers 3003, which is PTFE-based, in some designs. Non-PTFE materi- als are obviously a lot easier to work with than PTFE, so it's happening. Shaughnessy: How does the choice of material affect your layout process? Moyer: It has no effect on how you lay out the board. Before you get to layout, you build what's called the stackup where you say this is the material I want. is is its Dk and Df. And to Ed's point, are you using a traditional cop- per or a low profile, ultra-low, or very, very low-profile copper? Are you doing straight edges? Are you allowing for the undercut, the tapering effect, and so on? at is all done in a stackup before you ever place your first trace. is type of thing must be really well docu- mented on the old school paper master draw- ing. Now, with EDA tools, I've got the modern tools, algorithms, and metrics that will let me build an entire table that says, "Here's the con- struction, here's the material and the manu- facturer," providing the detail rather than the values. So you need to properly document what your assumptions are when you're try- ing to do this. Shaughnessy: So, would this change your constraints in any way? Moyer: is new information doesn't change the constraints. It's different mathematical data that goes into the same constraint. It's just different numerical values that are being used in the 2D field solver soware to calculate your characteristic impedance. Ed Kelley