Issue link: https://iconnect007.uberflip.com/i/972396
34 FLEX007 MAGAZINE I APRIL 2018 Finstad: Right, I think a lot of that gets up into the super-high frequencies. You start looking at 10 gigs and higher, that's where you get some people that are really sensitive to that. I've got one right now where I said, "Are you running anything with high speed?" and they were looking more in the 10+ gig range. They said, "We can't have the polyimide and the thermosetting adhesive properties at 10 gigs, even though the transmissions lines are very short." Sometimes I wonder if it's an electrical engi - neer who's looking too much at the numbers and not at real life, because anyone who's designed flex knows that that's your whole life. If you get the electrical engineer and the mechani - cal engineer together in the same room, they're speaking opposite languages. Every- thing that makes a flex better electrically makes it worse mechanically and vice versa. And there's always a com - promise. You rarely get every- thing you want. If you make it perfect and pristine from an electrical standpoint, but as soon as you bend it your conductors crack, it doesn't do you any good. There's always that balancing act and most of the time both sides have a little bit of give and take. They can handle a little bit less in impedance to get a little bit thinner, so you can still bend it; you may not get as thin as you want but it will be acceptable. There's always that balance on what you need, and sometimes I wonder if some peo - ple designing the front-end stuff, they're so wrapped up in just the numbers they don't realize that you probably could get by with something that doesn't perform at the level of an LCP and it will work almost as well. You get 90% of what you needed and people can actually build it. Shaughnessy: Do they over-constrain it because they're worried about it? Finstad: Yes, unfortunately a lot of engineers come out of college and they don't know squat about printed circuits. No one teaches flex cir- cuits in college. You learn electrical theory. All you've got are a bunch of numbers and you're learning the leading-edge stuff there and so when you get dumped in the industry, it's kind of the school of hard knocks. That's how you figure things out, by learning what actually works and what doesn't, and getting push-back from manufacturers on what can and can't be done. Shaughnessy: Scott or Scott? Either of you want to chime in here? McCurdy: Well, I've got a couple of things. I spent two days last week at the CES show in Vegas. And you walk around that 2.7 million square feet and you see it all. You see a glimpse of the future, and I can see where the flex business is going to have way more applications in the future in a lot of areas like robotics and things, and certainly automotive is going to suck up so many different interconnect things. And com - ing from the design world now, I think the designers need to kind of come to grips with maybe they haven't had flex experience before but I think there's more coming. So being able to understand more about the design for manufacturability. It's not just a rectangle anymore that you put six up on an 18 x 24 panel. They don't understand the panelizations and the fact that when they get to manufacturing, they're not rigid. You have to be thinking about different arrays, differ - ent ways of interlocking boards to get the maximum amount of panel. I mean, there's just a lot of that that is just kind of out of the norm of what most circuit board designers are used to. The designers aren't necessarily choosing the materials, but they're trying to make sure Scott McCurdy